Thunder and Herds: Rock Art of the High Plains 1598741527, 9781598741520

Citation preview

T HUNDER

AND

H ERDS

Loendorf ’s book on the rock art of the High Plains is a well-informed, vivid and highly readable presentation of one of the great rock art regions in the Southwest. One could not agree more with the author’s opinion that rock art sites are first and foremost archaeological sites and that ethnographic sources—together with archaeology—provide invaluable information on the rock art. This book is a must read both for the archaeological community and for the interested general public. —Jean Clottes, former director of prehistoric antiquities for the Midi-Pyrénées region of France; internationally acclaimed expert on painted cave art

If you thought the High Plains lacked rocks and thus rock art, you were wrong. Larry Loendorf ’s Thunder and Herds is the first major synthesis of an important but previously little known corpus of rock art, and it changes our understanding of the prehistory of the High Plains. It is essential reading for all North American archaeologists and rock art researchers. David S. Whitley, author of The Art of the Shaman: Rock Art of California and Introduction to Rock Art Research, and editor of The Handbook of Rock Art Research.

T HUNDER

AND

H ERDS

Rock Art of the High Plains

Lawrence L. Loendorf

First published 2008 by Left Coast Press, Inc. Published 2016 by Routledge 2 Park Square, Milton Park, Abingdon, Oxon OX14 4RN 711 Third Avenue, New York, NY 10017, USA Routledge is an imprint of the Taylor & Francis Group, an informa business Copyright © 2008 Taylor & Francis All rights reserved. No part of this book may be reprinted or reproduced or utilised in any form or by any electronic, mechanical, or other means, now known or hereafter invented, including photocopying and recording, or in any information storage or retrieval system, without permission in writing from the publishers. Notice: Product or corporate names may be trademarks or registered trademarks, and are used only for identification and explanation without intent to infringe. Library of Congress Cataloguing-in-Publication Data Loendorf, Lawrence L. Thunder and herds : rock art of the High Plains / Lawrence L. Loendorf. p. cm. Includes bibliographical references and index. ISBN 978-1-59874-152-0 (hardback : alk. paper) — ISBN 978-1-59874-151-3 (pbk. : alk. paper) 1. Indians of North America—Great Plains—Antiquities. 2. Petroglyphs—Great Plains. 3. Rock paintings—Great Plains. 4. Picture-writing—Great Plains. 5. Great Plains—Antiquities. I. Title. E78.G73L64 2008 978'.01—dc22 200802826 ISBN 978-1-59874-152-0 hardcover ISBN 978-1-59874-151-3 paperback

C ONTENTS

LIST OF ILLUSTRATIONS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6 LIST OF TABLES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9 ACKNOWLEDGMENTS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11 PROLOGUE—Thunder and Herds: Rock Art of the High Plains . . . . . . . . . .14 CHAPTER 1 Foreground and Background: The Physiography, Rock Art Research History, and Cultural Chronology of the Central High Plains . . . . . . . . .19 CHAPTER 2 Rock Art of the Archaic Period . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45 CHAPTER 3 Purgatoire Pecked-I Petroglyphs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 81 CHAPTER 4 Purgatoire Pecked-II Petroglyphs. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 110 CHAPTER 5 Purgatoire Painted Rock Art of the High Plains . . . . . . . . . . . . . . . . . . 142 CHAPTER 6 Rock Art of the Protohistoric Period: A.D. 1450–1725 . . . . . . . . . . . . 165 CHAPTER 7 Incised and Painted Rock Art of the Historic Period. . . . . . . . . . . . . . . 190 CHAPTER 8 Through a Glass, Darkly . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 220 REFERENCES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 232 INDEX . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 244 ABOUT THE AUTHOR . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 254

L IST

OF

I LLUSTRATIONS

P-1 1.1 1.2 1.3 1.4 1.5 1.6 1.7

A rock art researcher with “tunnel vision.” . . . . . . . . . . . . . . . . . . . . . . . . . . . .17 Map of the Central High Plains . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .20 Photograph of a typical High Plains vista . . . . . . . . . . . . . . . . . . . . . . . . . . . . .21 Overview of Purgatoire (Picket Wire) Canyon . . . . . . . . . . . . . . . . . . . . . . . . .22 A view from the top of one of the many dikes in southeastern Colorado . . . . .24 Etienne Renaud at site 5PE62 in Turkey Creek, Fort Carson, Colorado . . . . .26 Anne Whitfield at a site in southeastern Colorado . . . . . . . . . . . . . . . . . . . . . .30 A seriation of sites from the Middle Archaic through the Diversification Periods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .37

2.1

2.16 2.17

Close-up photograph of exposed Clay Creek petroglyphs in the Dakota sandstone formation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .47 The middle section of the Clay Creek petroglyph panel . . . . . . . . . . . . . . . . .49 A view of 5LA5598, one of the Piñon Canyon boulder sites . . . . . . . . . . . . . .52 Selected petroglyph boulders from the boulder sites at the Piñon Canyon Maneuver Site . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .53 Quadruped petroglyphs from the Piñon Canyon boulder sites . . . . . . . . . . . .54 A pit feature at 5LA5598, one of the Piñon Canyon boulder sites . . . . . . . . . .55 Petroglyphs on the exposed sandstone surface at Glorieta Mesa . . . . . . . . . . . .58 One of several excavations to expose buried petroglyphs at Glorieta Mesa . . . .59 Scale drawings of petroglyphs at sites on Glorieta Mesa . . . . . . . . . . . . . . . . . .65 Scale drawings of petroglyphs at sites on Glorieta Mesa . . . . . . . . . . . . . . . . . .66 Abstract petroglyphs from the Hicklin Springs site in southeastern Colorado . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .69 Petroglyphs from 5LA9781, a site near the Bent Canyon Stage Stop, Piñon Canyon . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .70 Quadruped figure on a basalt boulder at the Hogback Archaic site, Piñon Canyon Maneuver Site . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .71 Vertical figures at the Hogback Archaic site, Piñon Canyon Maneuver Site . . .72 The recording of one of the petroglyph boulders at the Ancient Hogback site, Piñon Canyon Maneuver Site . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .74 A tracing of the incised figure on Boulder 1 at the Ancient Hogback site . . . .75 A tracing of the incised figures underlying abstract pecked images . . . . . . . . . .76

3.1

A few of the quadruped figures on a boulder at the Bull Pasture site . . . . . . . .84

2.2 2.3 2.4 2.5 2.6 2.7 2.8 2.9 2.10 2.11 2.12 2.13 2.14 2.15

3.2 3.3 3.4 3.5 3.6 3.7 3.8 3.9 3.10 3.11 3.12 3.13 4.1 4.2 4.3 4.4 4.5 4.6 4.7 4.8 4.9 4.10 4.11 4.12 4.13 4.14 4.15 4.16 4.17 4.18 5.1 5.2 5.3

An anthropomorph at the Bull Pasture site holding a hoop toward a quadruped . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .86 A portion of the rock art panel at the Zookeeper site . . . . . . . . . . . . . . . . . . . .89 A loop-line figure that is connected to several of the quadrupeds at the Zookeeper site . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .90 A view of the Point site, located on the prominent sandstone outcrop, as seen from the Zookeeper site . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .91 The sandstone slab-wall structures at the Point site . . . . . . . . . . . . . . . . . . . . .92 The Big Hands Hunter petroglyph . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .96 One of the bubble-head quadrupeds at the Big Hands Hunter site . . . . . . . . .97 Quadruped petroglyphs from the Big Hands Hunter site . . . . . . . . . . . . . . . .98 Quadruped figures and intersecting spears at the Big Hands Hunter site . . . . .99 Anthropomorph petroglyph from the Big Hands Hunter site . . . . . . . . . . . .100 Human footprints at the Big Hands Hunter site . . . . . . . . . . . . . . . . . . . . . .100 Quadruped figure with a net-like head at the Big Hands Hunter site . . . . . .102 An overview of the Red-tail Rockshelter . . . . . . . . . . . . . . . . . . . . . . . . . . . .113 Quadrupeds with interlocking horns at the Red-tail Rockshelter . . . . . . . . . .114 A red-painted anthropomorph at the Red-tail Rockshelter . . . . . . . . . . . . . . .115 A group of quadrupeds at the Red-tail site shown running to and from a net . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .115 Quadrupeds caught in a net at the Red-tail site . . . . . . . . . . . . . . . . . . . . . . .116 Strike marks on a painted quadruped at the Red-tail site . . . . . . . . . . . . . . . .117 A deep bedrock metate at the base of a petroglyph panel at the Red-tail site . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .118 A starburst design painted with yellow pigment on the ceiling of the Red-tail site . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .121 The major petroform at site 5LA5555 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .123 Dancing figures at the Petroform site and similar figures from nearby site 5LA5549 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .124 The petroglyph panel at the Corral site . . . . . . . . . . . . . . . . . . . . . . . . . . . . .127 A map showing the relationship of the features at the Corral site . . . . . . . . .128 Archaeologist Mark Owens with the auger probe at the base of the Corral site panel . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .130 A single boulder at site 5LA5589 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .132 A single boulder in the Purgatoire River bottom, with a hunting scene on its sides and top . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .132 A bear and an unidentified companion figure at site 5LA5589 . . . . . . . . . . .133 A very rare example of a Diversification Period hunter with a bow at site 5LA296 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .133 An anthropomorph at site 5LA5589 whose hand is connected to a spear embedded in a quadruped . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .134 The Game Drive site . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .145 A typical panel of pecked and painted figures at the Game Drive site . . . . . .146 A painted anthropomorph and quadruped at the Game Drive site . . . . . . . .147

5.4 5.5 5.6 5.7 5.8

A contour map of the Game Drive site . . . . . . . . . . . . . . . . . . . . . . . . . . . . .149 The loop-line and net-like figures at the outlet of the Game Drive site . . . . .150 Abstract pictographs above the opening of the Rock Crossing rockshelter . . .153 The pecked quadruped and a turkey-like bird at the Rock Crossing site . . . .153 A view of the Bear Dance site showing the rock wall that encloses the rockshelter . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .155 5.9 A plan-view map of the Bear Dance rockshelter site .......................................155 5.10 A scale drawing of the Bear Dance panel . . . . . . . . . . . . . . . . . . . . . . . . . . . .156 5.11 The dance scene in the Bear Dance panel . . . . . . . . . . . . . . . . . . . . . . . . . . .157 5.12 The deer, turkey and parfleche-like design from the Renaud rockshelter, Turkey Creek, Fort Carson, Colorado . . . . . . . . . . . . . . . . . . . . .163 6.1 6.2 6.3 6.4 6.5 6.6 6.7 6.8 6.9 6.10 6.11 6.12 6.13 6.14 6.15 6.16 7.1 7.2 7.3 7.4 7.5

7.6 7.7

The Segesser I hide painting . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .168 A contour map of the Sue Site showing the locations of the excavation units and the main petroglyph panel . . . . . . . . . . . . . . . . . . . . . .169 The main petroglyph panel at the Sue site . . . . . . . . . . . . . . . . . . . . . . . . . . .171 A bison petroglyph at the Sue site . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .172 A field drawing of a bison at the Sue site . . . . . . . . . . . . . . . . . . . . . . . . . . . .172 Representative bison and deer from the Stone Structure site . . . . . . . . . . . . .174 A pit structure in the Smith dike that is very similar to features at the Stone Structure site . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .177 An anthropomorph at the Smith Dike that appears to represent a lightning gans . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .177 A bison with heart line from the Dinetah compared to a similar bison from southeastern Colorado . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .179 The turtle-like figure at the Stone Structure site and a sand painting of a turtle . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .179 An example of an Apache figure carrying a sinew-backed bow . . . . . . . . . . . .180 Petroglyphs representative of the Apache anthropomorphs in the region . . . .181 Gan dancers from the Piñon Canyon Hogback . . . . . . . . . . . . . . . . . . . . . . .182 Rio Grande–style warriors in profile view carrying bows and hatchets . . . . . .185 A Rio Grande style figure in bas-relief that may adorn a shield . . . . . . . . . . .186 The figure of a bison superimposed over an older Apishapa-age pecked quadruped . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .187 A stereo photograph of the Purgatoire bear taken by B. H. Gurnsey about 1889 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .194 A large painted bear at the Red Top Ranch, Colorado . . . . . . . . . . . . . . . . . .195 Oversize figures from the Hicklin Springs site in southeastern Colorado . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .196 The transformation scene at Hicklin Springs . . . . . . . . . . . . . . . . . . . . . . . . .197 Composite illustration showing the anthropomorph at Picture Canyon that may be suffering from smallpox and the Battiste Good drawings of smallpox and the Aztec pain symbol . . . . . . . . . .200 Anthropomorph figure at Picture Canyon surrounded by cupules . . . . . . . . .201 A drawing of the two-lance figure at Picture Canyon. . . . . . . . . . . . . . . . . . .203

7.8 7.9 7.10 7.11 7.12 7.13 7.14 7.15 8.1 8.2

A human figure standing on the back of a horse at the Picture Canyon site . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .204 A ribstone at Picture Canyon . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .205 The abstract pattern in Crack Cave, southeastern Colorado . . . . . . . . . . . . .208 Typical Biographic image of horse and rider. The figure is at the Cairn site on the Piñon Canyon Hogback . . . . . . . . . . . . . . . . . . . . . . . . . . .209 Two illustrations of a set of red-painted guns at the Red Guns site . . . . . . . .210 A drawing of the pecked horse and rider at the Red Guns site . . . . . . . . . . . .211 A small incised horse with standing rider at the Red Guns site . . . . . . . . . . .212 A ribstone on a detached fragment of the canyon wall at the Picnic site, southeastern Colorado . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .215 Concentric circles and other branching motifs at the Philmont Scout Ranch . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .224 Suspected Apache painted figures that are often confused with Purgatoire Painted images . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .229

L IST

1.1 1.2 1.3

OF

TABLES

High Plains Cultural Sequences . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31 High Plains Seriation Categories and Motifs . . . . . . . . . . . . . . . . . . . . . . . . . . 35 Changes in Quadruped Attributes through Time. . . . . . . . . . . . . . . . . . . . . . . 40

This page intentionally left blank

A CKNOWLEDGMENTS

y interest in southeastern Colorado rock art began nearly two decades ago when my wife Paula and I, guided by Steve Chomko, visited petroglyph sites on the Pinon Canyon Maneuver site.1 After seeing hundreds of watermelon-sized-boulders in Van Bremer Arroyo with petroglyphs on their tops and sides, and trekking onto the Hogback to see more sites, I knew I wanted to focus my research on the region's rock art. In subsequent years I have been fortunate to have had the opportunity to record many already known sites and to discover dozens of new ones along the Hogback and in the canyons of the Purgatoire River drainage. I could not have done this research without the support of many other individuals. Foremost, I want to thank Tom Warren, Director of the Directorate of Environmental Compliance and Management (DECAM), and Mary Barber, Deputy, DECAM, for their support throughout the years I have worked at the Pinon Canyon Maneuver site. Tom and Mary are true advocates for the protection of archaeological and historical sites at Fort Carson and the PCMS. Steve Chomko and Randy Korgel, former administrators of DECAM's cultural resources program, also supported rock art research. More recently, encouragement on the part of Pamela Cowen, Cultural Resources Manager, DECAM, has led the Pinon Canyon rock art projects to discover necessary funding sources. Pamela's support has also allowed me to make use of volunteers to complete the Hogback survey. Fort Carson archaeologist Mark Owens and his wife, Pamela, have been particularly helpful with recording and protecting rock art in southeastern Colorado. Terry Moody has been a considerable help with finding illustrations at the Fort Carson Curation Facility as well as with locating and recording rock art sites. I'm also grateful for her work as a volunteer on a half-dozen projects in the past decade. Thad Swan made exceptionally good maps of the Game Drive site

M

12

L AWRENCE L. LOENDORF — THUNDER

AND

HERDS

and the Corral site. Kendra Rodgers found and recorded the Bear Dance site and helped with several other rock art projects. Other important Fort Carson personnel include Mead Klaveter and Seija Karki, two biologists who have become converts to the protection of rock art sites. I am indebted to the ranch owners and managers who have allowed me access to sites on the properties they own or maintain. These include Brett Bannon, Harold Daniels, Richard Hammer, Roger Long, Mary Moore, Excel Smith, Jerry Wenger and other ranchers who wish to remain unnamed. I appreciate the assistance of Santa Fe Forest archaeologist Brent Abel, who showed me sites at Glorieta Mesa, and archaeologist Michelle Stevens, who gave me access to Comanche Grasslands site records. I was treated with considerable kindness by Susan Collins, Todd McMahon and other staff at the Colorado Office of Archaeology and Historic Preservation. The site records and photographs of Daphne and John Rudolph were very helpful, as were the recording forms compiled by Ken Andresen for the Hicklin Springs site. Field drawings by Peter Faris of panels at Hicklin are especially good. Thanks also to Bill Doleman and the staff at the New Mexico Archaeological Records Management System. Max Canestorp read the chapter containing a discussion of the Zookeeper site and told me how the site was actually discovered. My longtime friend Robin Beaver read and offered creative suggestions for sections of the book. Frank Schambach sent me information on the Crenshaw site in Arkansas. Bonita Newman helped with illustrations and site recording, and did the research using the portable x-ray fluorescence instrument. Dennis Slifer guided me to sites and sent me information about others. I am very thankful for the help of local rock art enthusiasts Mike Waugh in Colorado and Nancy Robertson in New Mexico, both of whom took me to interesting sites. John Robertson, Nancy's brother-in-law, spent a long day showing me sites in the Eagle Tail area. In recent years, I have appreciated Ann Whitfield's knowledge about regional rock art and her generosity in sharing it. Linda Olson, Minot State University, is an accomplished recorder of rock art sites. She and her crews have made remarkably good records of regional sites like Red-tail Rockshelter and Big Hands Hunter, both of which are featured in this book. Other field crew members have helped with rock art recording projects, and I would like to thank them for their time-Joe Artz, Kelli Barnes, Evelyn Billo and Robert Mark, Mike Bies, Stuart Conner, Cheryl Damon, Jackie Duncan, Dee Dunn, Ray Freeze and Carol Georgopoulos, Mike Gregg, Kerry Hackett, Peter Halter, Ann Hayes, Cherie Howey, Cynthia Kordecki, Dave Kuehn, Terry and Laurie Lee, Chris Loendorf, Janet Lever-Woods, Elizabeth Lynch, Christine Martinez, Castle McLaughlin, Belinda Mollard, Don Nordstrom (deceased), Brad O'Grosky, Christine O'Toole, Claudette Piper, Steve Rathman, Kathryn

ACKNOWLEDGMENTS

13

Rand, Lloyde Richmond Jr. (deceased), Lisa Shifrin, Mike Waugh, Amanda Winchter, and Ann Whitfield. While working on the manuscript of this book, I have been ably assisted by several individuals. Foremost is the talented editor, Nancy Medaris Stone, who edited and improved every section of the book. I appreciate the help of Elaine Nimmo, who redrafted a number of the illustrations. I am also grateful for David Whitley's editorial overview of the manuscript and the suggestions he made to improve it. I thank Jean Clottes and Jim Keyser for reading the manuscript and offering to write promotional material. The comments of one anonymous reviewer were also helpful. I am especially grateful to Mitch Allen at Left Coast Press for his support in getting the book into print. Mitch has become a true advocate for the archaeological community across the world. Carol Leyba and Robert Leyba of Leyba Associates did a phenomenal job with the final edit, the book design and layout. I strongly recommend their work to anyone looking for a book designer. Finally, I want to thank my wife, Paula, for putting up with me during the extended time I have worked on High Plains rock art projects. I am also pleased by her willingness to accept that my work is not done and that there is much more I want to learn about the incredible rock art images on the High Plains.

NOTE 1. Two spellings of the name of the “Pinon Canyon Maneuver Site” occur in this book. Since the

U.S. Army does not use the tilde that is normally present when the Spanish word “piñon” is written, I have used the official governmental spelling of the site’s name when I acknowledge the help I have received from PCMS personnel. Elsewhere, I use the customary Spanish spelling of “piñon,” which includes the tilde.

PROLOGUE THUNDER ROCK ART

AND

OF THE

HERDS

HIGH PLAINS

or anyone who has lived in a High Plain’s field camp—as I have during much of the past 50 years—few things are more memorable than the violent summer thunderstorms that meteorologists call mesoscale convective systems. I and others in my profession who work on the High Plains have many vivid memories of the efficiency with which the driving winds and hail associated with a summer thunderstorm can flatten a field camp and terrify its occupants. The worst of these storms announces its intrusion into an otherwise lovely afternoon with the sound of distant thunder, which gradually becomes louder as the sky steadily darkens. The ominous advancing clouds are not as uniformly black as the back of an Angus bull in the noonday sun, but are instead interlaced with white bands of ice balls. Since these hailstones can be golf ball size or larger, the apprehensive archaeologist can only hope they will be released somewhere other than over the field camp. Sometimes the black clouds will be streaked with gray, and if the winds are extremely violent, these stringy-looking bands of vapor will begin to swirl and rotate, creating a vortex. Contrary to popular apprehensions about tornadoes, however, the worst storms are only truly threatening when they begin to roll across the landscape like a massive mowing machine capable of stripping everything in its path. It is at this point that the experienced archaeologist realizes that most of his field camp may soon disappear. As the wind picks up, a frenzied effort to protect the camp begins. Tent pegs are reinforced with whatever rocks are nearby, gear is stashed in a pile behind any feature that may act as a windbreak, all the while sand and grit become airborne and attack any exposed part of the human body. The very fortunate who survive the onslaught relatively intact are not, as might be expected, jubilant. If by chance they had placed their field camp in a canyon bottom or on a nearby slope,

F

PROLOGUE

15

they are about to discover that the true fury of these storms arrives in the form of torrents of water that has fallen on the flat uplands above. Relatively quickly the currents will flow into a channel or arroyo that drains into a canyon—their canyon—and will carry away to the next county whatever the winds haven’t already demolished. An archaeologist quickly learns that when buying a tent, the shopping list must include an extra set of poles. Another strategy I have used for years is to buy two inexpensive tents—one to set up and camp in and the second to keep in its box as insurance against disaster. In the unlucky event that the first tent is ripped to shreds by the wind, there is still a spare one to use as shelter for the remaining nights of the field season. Fortunately, not all thunderstorms have devastating consequences. Meteorologically more normal events, usually occurring weekly in the summer, begin with a darkened sky and lightning, lots of lightning. These storms are as prevalent on the High Plains today as they were when the rock art discussed in this book was made and, as the reader will learn, I believe that lightning and thunder played a significant role in the placement of some rock art sites and on the choice of subject matter inscribed there. This is particularly true of the basalt dikes that crisscross the grasslands along the Colorado and New Mexico border east of Spanish Peaks and appear to have been magnet locations for both lightning and rock art. The herds of big game animals that inhabited the region’s open steppes had an equally formative effect on the character of the rock art imagery. Buffalo, antelope, mule deer, whitetail deer, elk, and bighorn sheep were all abundant on the High Plains, and, since they played such an important role in human subsistence, it is not surprising that all of these species are depicted in the rock art of their human hunters. In the pages that follow, the reader will learn that images of these herd animals frequently occur in groups that are associated with small human figures who appear to be driving them into nets or traps. I have called this book Thunder and Herds, a reworking of the title of Zane Grey’s novel The Thundering Herds which alludes to the thunder-like sound made by the pounding hooves of a buffalo herd in flight. Both elements—the thunder and the herds—were important constituents of life on the High Plains as experienced by prehistoric, rock-art-making inhabitants. The novel was made into two motion pictures, the best known directed by Harry Hathaway and released in 1933. In this version Randolph Scott portrayed a buffalo hunter who helped Plains Indians stop (at least temporarily) the decimation by hide hunters of local buffalo populations. Although the movie was filmed in the Lamar Valley of Yellowstone National Park, partly because that was the only locality with credible numbers of buffalo, the intent was to illustrate the size of the herds extant on the High Plains 100 years earlier.

16

L AWRENCE L. LOENDORF — THUNDER

AND

HERDS

Because I have worked for many years at the Piñon Canyon Maneuver Site in southeastern Colorado, the emphasis in this book is on the region that surrounds that facility and its extensive body of rock art. My team and I have done in-depth recording of several dozen rock art localities at the Maneuver Site, building in the process a substantial collection of comparative information that I have used as a framework for analyzing the imagery and patterns of the rock art inventory of the High Plains region. I have discovered during my years as a rock art researcher that it is often necessary to visit a site in order to understand it. So, while I would have liked to say more about the rock art of western Kansas and the panhandles of Texas and Oklahoma—and I have had access to the records and photographs from a considerable number of sites in these states—both my firsthand knowledge of the rock art and my discussion of it in this book are limited. The judgments that archaeologists make about the meaning and purpose of the rock art they investigate depend largely on the ideas they bring with them to a site. For instance, some researchers focus on individual rock art images and compare them to rock art figures at another site. This method of rock art analysis—called the iconographic approach—is interesting, but not nearly as fruitful as an examination of a site within its physical context or setting. I call this fixation with the images at a site to the exclusion of its surroundings the blinder syndrome. A person who examines a rock art site through a camera lens but overlooks the site’s archaeological remains is ignoring a potentially valuable source of temporal, cultural, and behavioral information and therefore can be said to be wearing blinders (Figure P.1). The basic premise underpinning my research and reflected in this book is that rock art sites are first and foremost archaeological sites. They are indeed places where people left images on rocks, but the rock-art-making activities were embedded in patterns of living that often become apparent by an analysis of the artifacts and features discarded at the site. Sometimes the associated artifacts and features described in this book have been found through excavation, but in most instances they have been recovered from site surfaces. I have discovered that it is as instructive to analyze the chipped stone debitage and ceramics at a rock art site as it is to examine the same material at a site where rock art is not present. Artifacts such as obsidian flakes from New Mexico’s Polvadera Peak are telling us the same thing, whether they are found at a rock art site or on a lithic scatter site: that the individuals who left them were engaged in the same interaction spheres. A second premise underlying my research is that ethnographic sources can contribute significantly to an understanding of rock art sites. By the term ethnography, I mean more than written records. Items preserved by, actively used by, or depicted by traditional sources, such as apparel that has changed very little

PROLOGUE

17

through time, may also be represented in rock art. Many painted artifacts, such as robes, tipi covers, and ledger drawings, as well as early photographs, often contain clues to the contexts in which items depicted in rock art panels may have been used. My approach to rock art research and interpretation is not meant to diminish in any way the importance of rock art sites to American Indians. Indeed, for many Native peoples such sites are permeated with power because their elders

FIGURE P. 1. A rock art researcher with "tunnel vision" sees only the rock art and not the rest of the site. Illustration by Davíd Joaquín.

18

L AWRENCE L. LOENDORF — THUNDER

AND

HERDS

prayed there and their prayers survive in the surrounding rocks and soil. Because I am not part of this tradition, I have not received the spiritual preparation that would permit me to engage a rock art site from this perspective. My Native colleagues recognize, however, that even though I can only approach a site in terms of my own background and education, I support the right of those who are spiritually prepared in Native traditions to visit rock art sites for their own purposes. From the vantage point of my long experience as an archaeologist on the High Plains, the advice I would offer to a student who is on the threshold of a career in archaeology and wants to do rock art research is this: learn as much as you can about how to be an excellent field archaeologist; consider the entire archaeological record of a region to be relevant, because the remnants left behind by prehistoric humans reflect their use of landscapes and not just rock art sites; explore ethnographic sources for information about the lifeways, beliefs, and forms of graphic expression of historic-era tribal peoples; and resist the impulse to focus exclusively on an iconographic comparison of rock art figures. I also think that it would be a good idea to prepare for the worst by buying two tents and an extra set of tent poles.

1 FOREGROUND AND BACKGROUND T H E PH Y S I O G R A PH Y, RO C K A RT R E S E A RC H H I S TO RY, AND

C U LT U R A L C H RO N O L O G Y

THE

OF

CENTRAL HIGH PLAINS

he High Plains—for many, the name evokes scenes in Hollywood westerns of rolling tumbleweeds and blowing dirt. Others think of the High Plains as the endless flat landscape that must be crossed before travelers from the East arrive at the Rocky Mountains for their summer vacation. For some who live there today, the High Plains are a source of the nutritious short grass that feeds their herds of cattle. For the American Indians who lived there for millennia, the bountiful resources of the High Plains sustained life in all of its dimensions: nutritional, social, and spiritual. Rock art is a graphic form of expression left by the prehistoric peoples of the High Plains that celebrates their web of existence, and it is the subject of all subsequent chapters of this book. But before I can begin my discussion of the region’s rock art figures, the localities where they occur, and their possible meaning, there is other ground I must explore in this chapter, some physical and some intellectual.

T

19

20

L AWRENCE L. LOENDORF — THUNDER

THE PHYSIOGRAPHY

OF THE

AND

HERDS

CENTRAL HIGH PLAINS

The High Plains constitute an upland region that begins just east of the great Rocky Mountain chain and ends at the 100th meridian, another name for an imaginary line passing from the North to the South Poles and widely known in North America as “the place where the West begins.” Located in the rain-shadow of the mountains to their west, the High Plains are significantly drier than the tall grass prairies to their east.1 The rock art discussed in this book occurs in the portion of the High Plains between Colorado Springs, Colorado, and Wagon Mound, New Mexico (Figure 1.1), a region I often, although not always, refer to as “the central High Plains.” Compared with other parts of the High Plains, the physical features of this central area are less uniform, due to the geologic history and subsequent biotic diversity of the Platte River basin in the north, the Arkansas River basin in the middle, and the Canadian River basin in the south.

FIGURE 1.1. Map of the Central High Plains. Illustration by Davíd Joaquín.

1 — FOREGROUND

AND

BACKGROUND

21

In prehistoric times, millions of pronghorn, bison, deer, and elk—the big game animals needed to support a successful hunting economy—roamed this high, flat region (Figure 1.2). Before Europeans arrived, bringing the horse with them, it was difficult for a hunter on foot to capture these grazing species in any significant numbers. Then as now, in wide open terrain, game animals rely on their eyesight and sense of smell to detect predators, and they simply run away when hunters approach. To be successful in the pre-horse era, hunters searched for game in broken terrain—landscapes where ridges, hills, arroyos, and canyons provided protective cover and allowed them to approach animals stealthily, without being detected. Although such settings occur along river courses on the open plains, they are more common at the borders or edges of flatlands, where erosion and runoff have dissected and remodeled the landscape into animal- and hunter-friendly habitats. It is the juxtaposition of these features that offered prehistoric hunters optimal conditions for heavily exploiting faunal resources, a fact attested to by the archaeological evidence found there. These locations are also the places where rock formations are exposed and, not surprisingly, where hundreds of rock art sites were created. The canyons are perhaps the most spectacular landforms in the region. Purgatoire Canyon, known locally as the Picket Wire, reaches depths of 100 meters below the surrounding uplands (Figure 1.3). Intersecting canyons like Van Bremer Arroyo or the Chacuaco Creek Canyon, where water flows only

FIGURE 1.2. Hundreds of square kilometers of the High Plains are flat. Note the ridge of hills in the far distance in this photograph. Photograph by Lawrence Loendorf.

22

L AWRENCE L. LOENDORF — THUNDER

AND

HERDS

intermittently, are relatively shallow and dirt-sided along their upper reaches, but farther downstream they are transformed into twisting, narrow sandstone channels. Where the upper end of Van Bremer Arroyo traverses the Piñon Canyon Maneuver Site, it is easily crossed by vehicle or on foot, but it has been transformed into a steep-sided canyon that is more than 80 meters deep by the time it meets the Purgatoire. The slopes of arroyos and canyons provide good habitat for two species of juniper that offer nesting cover and food for songbirds. Species of willows and cottonwoods border the watercourses, where their roots can access subsurface moisture, and they provide good browse for whitetail and mule deer, bighorn sheep, as well as porcupines and rabbits. Roosting and nesting birds also find food and shelter in their foliage. Predators follow their prey, so it is not surprising that coyotes, bobcats, and mountain lions are also common in these dissected habitats. Black bears are found in the area today, and grizzly bears were frequent inhabitants in the past.2 With their rockshelters, caves, overhangs, and twists and turns, the canyons of the central High Plains provided ideal shelter and subsistence opportunities for prehistoric hunting peoples. A fascinating archaeological record and more than 300 rock art sites have already been recorded, and there is undoubtedly much more rock art remaining to be discovered.3 The Colorado Piedmont and the Raton Section are other physiographic features of the central High Plains with exposed surfaces suitable for rock art. The

FIGURE 1.3. Overview of Purgatoire (Picket Wire) Canyon. Photograph by Lawrence Loendorf.

1 — FOREGROUND

AND

BACKGROUND

23

uneven topography of the Piedmont, just west of the Plains, is defined by ridges, steep-sided bluffs, flat mesas, and conical hills. Its most conspicuous feature, the Palmer Divide, separates the Arkansas and South Platte Rivers. Cresting at an elevation of 7600 feet (2300 meters), the divide is approximately 1500 feet (455 meters) higher than Colorado Springs and more than 2000 feet (610 meters) higher than Denver.4 To the south, the Piedmont blends into the Raton Section, a landscape defined by volcanic activity last occurring 14 million years ago. During tectonic episodes in which magma was forced upward, the existing bedrock was fractured, and a series of fissures filled with molten rock was created below the earth’s surface.5 Geologists use the terms basalt dikes and stocks to describe the reservoirs of cooled magma that occur across southeastern Colorado. These long, steep-sided ridges extending for as much as 15 kilometers are a lesson in the earth’s geological history. Formed over eons during the same period of vulcanism that created the Spanish Peaks—the primary mountains in the Sangre de Cristo range to the west of the central High Plains—the subsurface cavities in the dikes were exposed by erosion, revealing spectacular formations that look like the backbones of ancient reptilian creatures, appearing and disappearing below the surface of the landscape. Igneous rock formations like the Hogback at the Piñon Canyon Maneuver Site are more resistant to erosion than the softer sandstones in the region, and the many large clusters of petroglyphs at such locations undoubtedly owe their survival to the density of the bedrock. Although basalt dikes can be as much as 10 meters high in some places, their protruding porcine ridgetops can be less than a meter wide (Figure 1.4). Walking along the tops of these narrow ledges requires courage, not just because of their height and steep sides, but because the once molten basalt has fractured into blocks or chunks as it has weathered. A climber often discovers cavities where blocks have dislodged, creating an uneven path that is nerve-racking to negotiate. The missing blocks have tumbled down the dike’s sloping sides; some of the larger blocks roll to the bottom of the ridge, while others stop part way down, forming barrier walls. Another force of nature, lightning, has had a formative effect on the structure and character of these dikes. National Weather Service statistics reveal that the number of cloud-to-ground lightning strikes in a recent five-year period in Colorado counties is highest in Las Animas County, the location of the Piñon Canyon Maneuver Site. Its total of 35,800 strikes per year is greater than in any of Colorado’s other 63 counties,6 25 percent higher than in El Paso County, which had the second highest level, and 80 percent higher than the statewide county average of 6610 strikes per year. While there are undoubtedly many factors responsible for the meteorological conditions producing lightning, one that is pertinent here is the fact that the high iron content in basalt dikes such as the

24

L AWRENCE L. LOENDORF — THUNDER

AND

HERDS

FIGURE 1.4. Looking down from the top of one of the many dikes in southeastern Colorado. Photograph by Lawrence Loendorf.

Hogback acts as a magnet. The electrical charges striking the dikes have such force that they can fracture the basalt into heat-reddened shards. As might be imagined, the susceptibility of the dikes to lightening strikes has had important cultural implications for the location of rock art sites.

THE CLIMATE

OF THE

CENTRAL HIGH PLAINS

The ecological characteristics of the High Plains are determined by another set of dynamic forces encapsulated in the words climate and weather. A locality’s climate is the result of the interaction of its place on the earth in terms of distance from the equator and the North and South Poles (latitude) and its relationship to geographical features such as mountains and coastlines. Altitude—height above or below sea level—is another conditioning variable. The climate on the central High Plains today can be characterized as temperate. There is considerable sunshine; in fact, approximately 300 days of the year are more sunny than cloudy. Invariably there is some kind of daily wind, usually resulting from cool mountain air flowing from the west and interacting with warmer air on the flats. During the warmer months, the winds are strongest

1 — FOREGROUND

AND

BACKGROUND

25

in the mid-morning and evening. Weather conditions during the winter are less amiable, especially when “blue northers” bring blizzard conditions and heavy snow may accumulate up to several meters deep. These snowstorms can present life-threatening challenges, especially for newborn ungulates in the spring, but in most instances temperatures improve within a day or two and the reemergence of the sun initiates the melting process. As I have noted, the other predictable, disruptive weather phenomena are thunderstorms. Although these result from weather patterns approaching from all directions, in mid-summer, thunderstorms most commonly arrive from the southwest, bringing lightning and heavy rain. Occasionally they cause flash flooding, as the normally dry arroyos fill with rushing water, making passage to the opposite bank impossible. Summers can be extremely hot, with sustained periods of high temperatures. During July and August it is not unusual for the temperature to reach 90 degrees Fahrenheit for 40 to 50 consecutive days, although sunset brings some relief and nighttime temperatures usually cool off to comfortable levels.

A Reasonable Question A reader might reasonably ask, What does knowledge of the landforms and biotic communities of the central High Plains contribute to an understanding of cultural expressions such as rock art? The answer to this question begins with intellectual developments in archaeology during the 1970s and 1980s, when archaeologists became aware of the emergent field of ecology and its stress on the interconnectedness of natural forces and biological species. Attempts to explain what happened in the past began to include the recognition that a network of ecological factors and biological agents strongly influenced not only what life forms, including humans, were present in specific biotic zones, but also how they behaved. In prehistoric times this interrelatedness operated on the High Plains as much as it did in the Valley of Mexico or in the eastern woodlands of the United States. Addressing questions about why rock art occurs on the basalt ridges and canyon walls of the High Plains, what that rock art reveals about the people who made it, and why they created it, must therefore begin with a consideration of the characteristics of the natural world in which prehistoric peoples were integrated.

ROCK ART RESEARCH

ON THE

CENTRAL HIGH PLAINS

Several summaries present the history of archaeological research on the central High Plains, but in none of them is rock art research given the attention it deserves. I hope to remedy this oversight by presenting here an overview of previous rock art research in the region.7 It seems intuitively obvious that rock art in the West would be most frequently found in the great swath of the Rocky

26

L AWRENCE L. LOENDORF — THUNDER

AND

HERDS

Mountains and intermountain basins that runs north-south through the region. After all, that is where the majority of exposed rocks are located. While it is true that many rock art sites are found in these regions, when the numbers are tallied—as any local rancher will attest—southeastern Colorado’s canyon country and northeastern New Mexico’s volcanic terrain contain the greatest concentration of rock art sites on the central High Plains. This fact was recognized as early as the 1930s by Etienne B. Renaud, a French professor of languages at the University of Denver, who was one of the most colorful and industrious of Colorado’s early archaeologists (Figure 1.5). Often accompanied by his students, Renaud roamed widely across the western mountain states from northern New Mexico to central Wyoming in search of sites.8 Two other adventurous spirits from Columbia University, Haldon (Hal) Chase and Robert Stegler, set off on the High Plains Archaeological Expedition in 1949. After establishing their first field camp near the Purgatoire River in Chacuaco Canyon, they went looking for a site to dig. According to their field notes, their discoveries included very rough roads (and continual flat tires), lots of mosquitoes, a few tipi rings, and several rock art sites that Chase photographed. The photographs and site locations are archived at the University of

FIGURE 1.5. Etienne Renaud at site 5PE62 in Turkey Creek, Fort Carson, Colorado. The two red deer figures are examples of the Purgatoire Painted style. Photograph courtesy of the University of Denver Museum of Anthropology.

1 — FOREGROUND

AND

BACKGROUND

27

Denver, where they are available for study. Chase and Stegler are better known for their excavation later that summer at the Snake Blakeslee site on the Apishapa River, where they also recorded the associated rock art panels. In 1951, Hal Chase took a teaching job at Trinidad State Junior College and found there a faculty that was supportive of his archaeological interests.9 Throughout the 1940s and 1950s, Norman W. Dondelinger, the head of the Natural Science Department at Trinidad State, and Robert M. Tatum, of the United States Naval Academy, interviewed landowners and artifact collectors in the area, hoping to locate more archaeological sites. These two researchers were primarily interested in non-rock-art archaeological sites and published the results of their efforts primarily in Southwestern Lore. But Tatum was so impressed by the region’s many well-made petroglyphs that he wrote a separate paper describing them.10 Galen Baker, an instructor in archaeology at Trinidad State Junior College, continued the local tradition of studying rock art and also documented other aspects of the archaeological record. In a paper presented at the 1964 annual meeting of the American Anthropological Society, Baker defined a tentative rock art chronology for southeastern Colorado, based on techniques of manufacture, rates of erosion, and differential varnish accumulation on selected petroglyphs. Unfortunately, the paper was never published, but it was cited by Robert Campbell, then a graduate student at the University of Colorado (CU). Since Campbell’s primary goal was to establish an archaeological taxonomy for southeastern Colorado, he dug deep trenches into such sites as Medina rockshelter in the Chacuaco drainage. Based on his excavations at this and dozens of other sites in the inaccessible canyon country, Campbell developed a detailed chronology of post-Archaic changes in material culture and also recorded the rock art he encountered during his fieldwork.11 During the 1960s, academically based archaeologists throughout the country were focused on the classification of artifacts and sites and the development of chronological sequences. Because a similar temporal ordering of rock art was not then possible, researchers tended to concentrate on non-rock-art cultural resources. Viewed against the background of the preoccupations of the times, Campbell’s inclusion of rock art in his research is all the more remarkable, although there were so many rock art sites in southeastern Colorado it would have been difficult for him to ignore them. University of Colorado anthropology professor David Breternitz was an important figure in High Plains rock art research in the 1960s and 1970s. Even though he was not an active rock art recorder, Breternitz’s strong support of student research was a factor in the successful efforts of Robert Campbell and CU students Robert Burton and William Buckles, who each wrote master’s theses on rock art. Burton completed an innovative study of the images at Dinosaur National Monument in the northwestern corner of the state which included

28

L AWRENCE L. LOENDORF — THUNDER

AND

HERDS

factor analysis—a statistical method that correlates similarities in a sample. Buckles’s more traditional study recorded and analyzed the rock art at Medicine Creek Cave in the Black Hills of eastern Wyoming, and he later incorporated rock art analysis in his dissertation research on Ute prehistory in Colorado’s Uncompahgre Plateau region.12 Fortunately for regional rock art studies, Buckles continued to do rock art research during his long career at the University of Southern Colorado in Pueblo. He was also a strong supporter of the Colorado Archaeological Society (CAS), a group of professional and avocational archaeologists who have worked together on field and laboratory projects in Colorado for the past 70 years.13 On many of these projects, CAS researchers worked side by side with government and university archaeologists, excavating sites where no rock art was present. At other sites where rock art occurred, such as Hackberry Springs, in Baca County, Colorado, CAS teams not only recorded the rock art, but tried to develop a tentative chronology.14 The Pueblo and Denver chapters of the Colorado Historical Society have long had an interest in recording regional rock art sites. Sometimes they have been involved in organized projects like the one led by Ken Andresen in 1994, which recorded the Hicklin Springs site in Bent County, and at other times members have worked independently. John and Daphne Rudolph, who visited dozens of sites deep in the canyons of southeastern Colorado, personify those in the loner category. In the course of their monumental effort, the Rudolphs took thousands of photographs of rock art images and later donated their large collection of color slides to the Colorado Historical Society, where it can be accessed by researchers. Interest in avocational rock art studies is often stimulated by an energetic researcher who attracts a local following. Such a figure in southeastern Colorado was Bill McGlone, a retired engineer living in La Junta who developed a deep interest in rock art. Early in his research he believed that some petroglyphs were expressions of an ancient writing system, but over time his interest shifted to rock art that related to solstice and equinox marking. While many of McGlone’s ideas are not consistent with those of professional archaeologists, his research was characterized by a commitment to the scientific method and a search for ways to validate (or invalidate) his findings.15 Prior to the establishment of the boundary separating southeastern Colorado and northeastern New Mexico, mobile hunter-gatherer groups moved through and exploited the resources of the very large region now located on both sides of the state borders. Nancy Robertson, a historian and rock art enthusiast in Raton, New Mexico, has an immense knowledge of the archaeological sites and clusters of rock art images created by prehistoric peoples in this area. Over a number of years, she has recorded dozens of sites, kept meticulous records, and developed innumerable friendships with the area’s ranchers on whose lands many of the sites are located. New Mexico archaeologist Joseph Winter relied extensively on

1 — FOREGROUND

AND

BACKGROUND

29

Robertson’s records when he directed a cultural resource survey of the Dry Cimarron River region in northeastern New Mexico.16 Despite the extensive knowledge base that had been developed about rock art in southeastern Colorado and northeastern New Mexico, until recently sites in these areas were barely mentioned in important regional overviews.17 The visibility of rock art resources increased once the United States Army purchased and consolidated the seven ranches that, since 1983, have made up the Piñon Canyon Maneuver Site. Archaeologists have found and recorded 150 rock art sites there,18 some containing several hundred petroglyphs. Initially, the University of Denver (DU), under the direction of southwestern rock art specialist Sally Cole, was responsible for cultural resources research at the Piñon Canyon Maneuver Site. Cole identified the basic rock art styles at the military facility and established a solid base for future research. In 1987–88, I led a team from the University of North Dakota that completed intensive recording and analysis at seven Piñon Canyon rock art sites. In the following year, another five sites were recorded in conjunction with test excavations to expose buried petroglyph panels. The goal of the combined research at Piñon Canyon was to completely record the sites and to date the petroglyphs. As described in more detail later in this chapter, relative and chronometric methods for estimating the age of the rock art provided the data used to establish a regional rock art chronology.19 A number of rock art projects have been undertaken in adjacent areas since research first began at Piñon Canyon. An important study by Peter Faris applied the Piñon Canyon chronology to other sites in southeastern Colorado and confirmed that its relevance was region-wide. Several recent projects, including a survey in the Comanche National Grasslands of the Purgatoire Canyon bottom, revealed large numbers of rock art sites.20 These include an excellent example of Rio Grande–style rock art and many other images. One site, described in the archaeological journal Antiquity by University of Colorado graduate student Mark Mitchell, contains depictions of horses wearing armor.21 In another important study, Ralph Hartley and Anne Vawser examined the relationships among the region’s rock art sites, rockshelters, architectural features, and grinding surfaces.22 Using a series of statistically derived polygons containing these variables, the authors determined that rock art is more frequently found in the Purgatoire River bottom in association with bedrock metates than on the uplands at the Piñon Canyon Maneuver Site. Their approach allows investigators to ask why this difference occurs in more than one locality and permits the formation of hypotheses based on an examination of sites like the Redtail Rockshelter (discussed in Chapter 4), where a spatial separation occurs between the rock art and an activity area where grinding took place in the past. I use a more traditional approach in this book, placing rock art in its temporal and spatial frameworks, not just considering the association between surface sites and architectural remains, ground stone tools, rock art, and rockshelters.

30

L AWRENCE L. LOENDORF — THUNDER

FIGURE 1.6. Anne Whitfield with pin flags for volunteers to place at rock art panels. Photograph courtesy of Cheryl Damon.

AND

HERDS

Anne Whitfield, a member of the recently formed Colorado Rock Art Association, which is a chapter of the Colorado Archaeological Society, has directed several field excursions to record rock art on the Comanche National Grasslands (Figure 1.6). Students have also begun to present papers and publish studies of rock art in the central High Plains, a development promising that more will soon be known about the region’s impressive rock art endowment.23

CULTURAL CHRONOLOGY OF THE CENTRAL HIGH PLAINS Over time, what was once a single regional classification of archaeological sites on the central High Plains has become less uniform. In a 1999 study of southeastern Colorado, a revised classification of sites in the Arkansas River basin was developed by archaeologists Steve Kalasz, Christian Zier, and Mark Mitchell.24 Since the majority of the rock art sites I discuss in this book occur within this region, I have chosen to use it in this book (Table 1.1). The system is sufficiently general to account for most of the variation in the archaeological record, and it has facilitated my efforts to place rock art into a cultural chronology. It is at this point that rock art book authors usually list the cultural historical divisions and subdivisions of their region of interest. In my experience, however, readers tend not to absorb a sequence of phases and dates—such as “the Apishapa (A.D. 1050–1450) and Sopris (A.D. 1500–1200) are subdivisions of the Diversification Period (A.D. 1050–1450)”—when this information is presented independent of a discussion of the rock art and the dirt archaeology associated with these cultural historical units. I have chosen instead to arrange my discussion in more-or-less chronological order and to embed, as an example, a discussion of the rock art of the Archaic Period within a broader review of the regional variation in Archaic lifeways as understood from site excavations. Just as

1 — FOREGROUND

AND

31

BACKGROUND

TABLE 1.1. HIGH PLAINS CULTURAL SEQUENCES Southeastern Colorado

North-Central New Mexico

Cultural Taxon

Temporal Range

Cultural Taxon

Temporal Range

Protohistoric Period

A.D. 1450–1725

Jicarilla Phase

A.D. 1750–1880

Diversification Period

A.D. 1050–1450

Cojo Phase

A.D. 1550–1750

— Apishapa Phase

A.D. 1050–1450

Cimarron Phase

A.D. 1250–1300

Ponil Phase

A.D. 1100–1250

Escritores Phase

A.D. 900–1100

Pedrogoso Phase

A.D. 700–900

— Sopris Phase

A.D. 1050–1200

Developmental Period

A.D. 100–1050

Vermejo Phase

A.D. 400–700

Late Archaic Period

1050 B.C.– A.D. 100

Late Archaic notched projectile points

1850 B.C.– A.D. 400

Middle Archaic Period

4050–1050 B.C.

San Jose Phase

3250–1850 B.C.

Early Archaic Period

6850–4050 B.C.

Bajada Phase

4850–3250 B.C.

Jay Phase

5550–4850 B.C.

Plano Period

8300–6850 B.C.

Plano Period

8300–5550 B.C.

Folsom Period

8900–8300 B.C.

Folsom Period

8900–8300 B.C.

Clovis Period

9550–8900 B. C.

Clovis Period

9550–8900 B.C.

Pre-Clovis Period

> 9550 B.C.

Pre-Clovis

> 9550 B.C.

dirt archaeologists increasingly include rock art studies as a complementary data set in their survey and excavation reports, I have reversed the focus and incorporated as much archaeological information as possible about each of the rock art sites that I describe in subsequent chapters.25 Nonetheless, I believe that it will be useful to present a brief cultural overview for readers who are unfamiliar with the region’s history. Prehistory begins on the High Plains at the end of the Pleistocene, when the glaciers in the mountains to the west retreat. During the Paleo-Indian stage, the earliest inhabitants hunted mammoths and large bison with atlatls and darts tipped with projectile points whose names—Clovis, Folsom, and Scottsbluff (see Table 1.1 for dates)—are well known. In fact, the original discoveries of these projectile points occurred at sites on the High Plains.

32

L AWRENCE L. LOENDORF — THUNDER

AND

HERDS

Following the Pleistocene period, the onset of very hot and dry conditions had a direct effect on the High Plains. The large Ice Age mammals, the mainstay of Paleo-Indians, began to die off, and the diet of peoples depending on these species changed to a variety of plant foods, supplemented by smaller game animals. The Archaic stage, divided into Early, Middle, and Late segments, is the long period that designates this changing lifeway. The hot climate in the Early Archaic was particularly difficult for High Plains people, and the region was largely abandoned in favor of cooler and more humid locations in nearby mountains. During the Middle Archaic, groups returned to the region, living in the many rockshelters along the canyons and arroyos. Their sites contain large numbers of manos and metates which were used to grind dozens of different plant foods. By the Late Archaic, increasing numbers of deer and antelope bones in shelter deposits suggest a greater reliance on hunting. In the Developmental Period of the Late Prehistoric stage, open-air sites increased as people built small, rock-walled houses on the canyon rims or along arroyos. Both the bow and arrow and ceramics were introduced during this period. As the name implies, the subsequent Diversification Period was a time of regional specialization. During the Apishapa phase, population increased along the major rivers, where corn and beans were grown in small garden plots. Small villages could include as many as 25 to 30 houses made of dry-laid sandstone slabs with brush roofs. About 500 years ago the inhabitants of these villages are believed to have moved east, becoming identified as the Pawnee Indians of Kansas and Nebraska. Apache groups moved onto the High Plains in the Protohistoric Period, living in hide-covered tipis. After the arrival of the Spanish and the introduction of the horse, the region became home to the Comanche, Cheyenne, Kiowa, and other nomadic tribes. Each of these cultural entities, from Middle Archaic groups to Historic-era tribes, made their own distinctive rock art.

A Glossary of Pertinent Terms Like many research specialties, the study of rock art operates with a vocabulary that needs to be defined. While it is obvious that a rock art site is a “place where rock art is found,” the conventions used to define the site’s boundaries can vary from one investigator to another. Since much of my research has been conducted on the Piñon Canyon Maneuver Site, the site definition I use for cultural remains has been developed for that locality: “a cultural resource site is the location of past human activity with physical traces of that activity in the form of artifacts or features.” More specifically, a prehistoric site is “a locality with five or more unmodified chipped stone flakes or a single tool associated with one or more unmodified flakes distributed in such a manner that no artifact is more

1 — FOREGROUND

AND

BACKGROUND

33

than 20 meters from the next nearest artifact.” A location with a known buried component also qualifies as a site regardless of the number of surface artifacts that may be present. Similarly, any surface feature or prehistoric rock art locality is considered to be a site, whether or not it is associated with artifacts. Clearly, the preceding definitions are extremely restrictive, and their use results in the creation of a larger number of sites than would a less inclusive set of criteria. On the other hand, it must be remembered that surface artifacts are included in the definition, so while there may be a 20-meter interval between the rock art elements on a canyon wall, there are often surface artifacts and features connecting one group of rock art figures with its neighbors. In these situations, the site definition criteria can lead to the identification of such unmanageably large sites that archaeologists use naturally occurring boundaries, like a stream or the edge of an arroyo, to limit the size of their sites. The definition of a rock art panel is more straightforward. Panels are “rock surfaces containing images that are oriented, for the most part, in one direction.” The definition of a rock cleavage plane is, however, more technical: “When the direction or plane of the surface changes significantly, the surface is considered a new panel.” This criterion can become confusing when an image extends around the corner formed by the meeting of two sides of a rock or when images cover the entire surface of a rounded boulder. These problems are usually dealt with in the field by creating schematic sketches of a panel’s orientation.26 The most ambiguous term used in rock art research is style. For some archaeologists, the word simply refers to the attributes or general characteristics of the rock art they study. I use the term more restrictively to designate a group of similar images and motifs that are found within a recognized geographical region and a specific time period. My definition of rock art style is, therefore, the equivalent of an archaeological phase.27

Dating Rock Art One of the initial goals of archaeological research is to identify the contemporaneous artifacts, features, and sites in a region. Once temporal control has been established, patterns of stasis and change through time can be identified, and the question of why such patterns exist can be addressed. Establishing chronological control can involve two different kinds of analysis: if material remains from the past can be dated using physical processes, then the effort to establish a temporal sequence is accelerated. If, on the other hand, few appropriate techniques are available to date some classes of artifacts, such as petroglyphs, then sequencing usually begins with a process of intellectual housekeeping referred to as classification. Temporal control of the rock art on the central High Plains began with the development of a structure of categories for classifying the petroglyphs at the Piñon

34

L AWRENCE L. LOENDORF — THUNDER

AND

HERDS

Canyon boulder sites.28 The assignment of petroglyphs to specific categories was based on a set of variables that included image size and the size of the panel in which figures occurred, the percentage of the panel surface covered by rock art, the depth of the petroglyphs, a relative estimate of the degree of varnish cover, and, of course, the shape or form of each petroglyph element. Written criteria developed for descriptive categories, such as circles or curvilinear meanders, were strictly applied, and once it was decided that the system accounted for all known variability, a series of figure types was established, at first primarily for abstract images. Workable typologies are useful for a variety of tasks. Foremost is the comparison of petroglyph types from one site to another as the first step in developing regional petroglyph and pictograph styles. Seriation is another useful procedure which takes place after the development of rigorous types. Seriation has been used to establish regional rock art chronologies in several parts of North America.29

SERIATION A basic assumption of seriation is that artifact types, in this case categories of rock art figures, have the equivalent of a life cycle. The frequencies of a type such as circular forms will be low at point-A in time, then gradually increase to maximum popularity at point-B, and begin to decrease as another type, introduced at point-C, begins its own ascending trajectory. As I have noted, written criteria for classifying individual petroglyph forms were established and followed as the elements were assigned to types. The completed typology provides a vocabulary for discussing and quantifying the petroglyphs at the sites and for comparing them, either by individual category or as percentages of types within the complete inventory, to the rock art record at other sites. In developing the original seriation, I based the larger curvilinear and rectilinear categories on Heizer and Baumhoff ’s research. I placed the circles and circular shapes of types A and C in a curvilinear category, and the square and rectangular forms of types F and H into a rectilinear category (Table 1.2).30 In recent years, Nevada and California researchers have challenged the accuracy of the curvilinear and rectilinear categories, but this does not invalidate the High Plains seriation, which is based on the increase in quadrupeds and anthropomorphs through time rather than a change in the abstract categories.31 To create the seriation, the petroglyphs at each site were assigned to a motif, placed in their respective types using the classification criteria, and converted to percentages of all of the motifs at the site (Figure 1.7). Frequency seriation has its limits, however. Meaningful patterns of change through time can only be produced in a relatively small geographic area whose cultural development has been homogeneous. Because archaeologists believe—

1 — FOREGROUND

AND

BACKGROUND

35

TABLE 1.2. HIGH PLAINS SERIATION CATEGORIES AND MOTIFS Seriation Category Curvilinear

Types Combined for Seriation

Motifs within Types

Type A

Circle, u-shape, curved line, wavy line, curvilinear meander, curvilinear meander with enclosure, connected circle, rayed circle, spiral, curving rake

Type C

Bisected circle, circled line, tailed circle, spokedcircle, and concentric circle

Amoebas

Type E

Amorphous pecked shapes

Rectilinear

Type F

Square/rectangle, bisected line, bisected rectangle, bisected rectangular grid, rectangular form, rectangular form with enclosure

Type H

Cross, straight line, parallel series, intersecting line, and rake with square form

Dots

Type D

Dot, grid of dots, and pattern of dots in abstract forms

Quadrupeds

Type J

Quadrupeds undifferentiated for the seriation

Anthropomorphs

Type K

Anthropomorphs including stick and more substantial figures

Other

Unassigned

Bird and ungulate tracks, and other forms not in the typology

based on the absence of any evidence of new populations in the region—that the interval between the Archaic and the Late Prehistoric periods on the central High Plains was a relatively stable time, the historical arrangements of data that seriation provides are likely to be accurate. Inconsistency is the greatest problem encountered in developing a rock art seriation. If more than one researcher is responsible for coding petroglyph figures, one person’s interpretation of “interconnected lines” might be classified by someone else as a “stick anthropomorph.” Even though the criteria for each element in my typology are unambiguous, bias has still affected the act of classification, particularly on those occasions when I was accompanied by other researchers whose views influenced mine. Another problem is related to the necessity to make changes or additions to the list of element categories in a typology. For example, as time passed I realized that the figures I had originally classified as “intersecting lines” were actually bird tracks and, as such, they had to be removed from the “rectilinear types” category in

36

L AWRENCE L. LOENDORF — THUNDER

AND

HERDS

the seriation and placed in the class designated as “other.” Reclassification was not a serious problem in my seriation because few elements were added or reassigned, but if a site (and therefore the sample size) is small, then the percentages of element types will change somewhat with the addition or deletion of only a few figures. Despite the preceding issues, the seriation of High Plains rock art traditions from the Archaic through the Late Prehistoric Periods provides an excellent overview of changes in element representation through time (see Figure 1.7). It is important to recognize that the seriation includes sites from the Middle Archaic though the Diversification Period. I placed sites into the regional chronology based on those that have relatively good temporal control, such as Zookeeper. Once a site like Zookeeper was seriated, I could use it to assign sites with similar percentages of figures to the chronology. The abstract forms overwhelmingly present in the Archaic Period consist primarily of curvilinear figures, although all sites also have substantial numbers of rectilinear motifs. In my initial seriation, based only on the Piñon Canyon rock art, the rectilinear figures increased somewhat as time passed. The larger regional seriation suggested, however, that the prevalence of rectilinear figures is localized to Piñon Canyon.32 From the Late Archaic through the Diversification Period, rectilinear forms include what we now suspect are representations of nets. In the discussions in the following chapters, I have included the presence of nets in the more refined division of quadruped and anthropomorphic figures at sites dating to the Developmental and Diversification Periods. A significant increase in quadrupeds is evident in the seriation beginning in the Developmental Period and culminating in the Diversification Period (Figure 1.7). It is also clear that anthropomorphs are more common in later time periods, and the few examples that occur at older sites appear to be later additions. At Boulder site 5LA5598, for example, two or three anthropomorphs were added to a site that originally consisted of abstract figures. While the seriation illustrates that in later time periods the High Plains rock art record is dominated by quadrupeds, many of them are so similar that placing them in precise chronological order is difficult. For this reason I have developed a set of quadruped attributes (types of head appendages, presence or absence of hooves, association with spears or nets) whose changing frequencies have made possible their accurate assignment to time periods.33

A Preliminary Strategy for Assigning Quadrupeds and Anthropomorphs to Time Periods Guided by a basic principle of science, which is to proceed from the known to the unknown, I begin my discussion of the patterning visible in the quadruped category by examining the figures at sites where distinctive Late Archaic Period abstract

FIGURE 1.7. A seriation of sites from the Middle Archaic through the Diversification Periods. Sites in the Diversification category with only a few rock art figures were excluded from the seriation

38

L AWRENCE L. LOENDORF — THUNDER

AND

HERDS

petroglyphs provide a temporal anchor. The Hogback Archaic site 5LA10286 and the Boulder site locality 5LA5602 are examples of places where many abstract figures and fewer numbers of quadrupeds occur. A good argument can be made that the quadruped petroglyphs and the Archaic Period abstract images at these sites are contemporaneous and, as such, the quadrupeds can be used as a chronological baseline in a comparison of quadrupeds at sites from more recent time periods. Considered as a group, Archaic quadrupeds share a number of properties. Their body shape is variable, and they have relatively long legs and pointed snouts. They often have long ears or horns that point upward from the tops of their heads, but none of the figures has branching antlers. In contrast, quadrupeds at sites from the later Developmental and Diversification Periods often have branching antlers. Based on these data, we might be tempted to generalize that the presence of branching antlers is an indication that a quadruped dates to a more recent time period. Unfortunately, there are significant exceptions to this rule of thumb. If, for example, petroglyphs of quadrupeds with branching antlers were meant to represent deer, and petroglyphs of quadrupeds without branching antlers were meant to represent antelope, then the presence or absence of branching antlers would be an indication of the species depicted but would not tell us anything about where the figures belong in a temporal sequence. Furthermore, since female deer do not have antlers, petroglyphs of quadrupeds lacking antlers might also represent does rather than bucks. Nonetheless, even though it is not possible to write a hard and fast rule about what the presence or absence of branching antlers tells us about the age of a petroglyph, the generalization is relevant to the quadruped figures at Piñon Canyon when it is one of a number of attributes used to evaluate a figure. We might ask whether systematic variations in other characteristics, such as body shape, leg shape and length, foot treatment, and tail type, might help researchers discriminate between quadruped petroglyphs of different ages. A cursory examination of any group of quadrupeds reveals that an attribute such as body contour—which could be round, rectangular, or boat-shaped—does not serve a discriminating function. Quadruped figures have a variety of shapes, and all of these may be found on a single rock art panel. Decisions about a figure’s shape category are subjective, and consistency in classifying them is difficult to maintain. The relationship between measures of body shape—such as length compared with height— might be useful but, so far, these ratios have not been easy to systematize. Quadruped tails have also been difficult to categorize, but for a different reason. Instead of sometimes having tails and sometimes not, or exhibiting differences in tail posture, rock art quadrupeds invariably have tails and most often they point straight up. Such uniformity does not help researchers discriminate among quadruped figures, but it does provide important information about animal behavior and supports a judgment that the animals are fleeing, since frightened deer and antelope run with their tails up.

1 — FOREGROUND

AND

BACKGROUND

39

Quadruped legs and feet offer more analytical potential than body shape and the presence and position of tails. Establishing a ratio between leg length and body height may eventually produce useful information, and the same is true for the ratio between a figure’s longest leg and its shortest. A large number of quadrupeds would have to be measured, however, to determine whether there is consistent variability that separates along typological lines. In contrast, meaningful variety in the feet of quadruped petroglyphs at Piñon Canyon has been documented. Although the legs of many figures terminate without feet, a significant number have inverted U-shaped hooves, and the feet of a few figures are round and bulbous. Interestingly, hoof treatment tends to be correlated with different kinds of head appendages. For example, none of the recorded quadrupeds that have elaborate or atypical branching antlers also have legs ending in feet. Temporal patterning in the kinds of images associated with quadrupeds is also apparent. The presence of spears, arrows, and nets, as well as contiguous or closely associated anthropomorphs, increases through time. I have used the preceding variables to make some preliminary chronological assignments of the quadrupeds at rock art sites on the Hogback, along Van Bremer Arroyo, and elsewhere at Piñon Canyon (see Table 1.3). In order to emphasize the criteria in the groupings that distinguish one chronological unit from the other three, those distinguishing attributes are indicated by italics. Since I have chosen to use a polythetic system of classification—in which the members of a class share some but not necessarily all of the defining characteristics of the class—anomalies or other kinds of exceptions not included in the classification criteria may be included in the groupings. In other words, this system is designed for researchers who would be called lumpers rather than splitters. It should also be noted that the criteria have been selected to work with large numbers of quadrupeds, either at the level of a panel or of an entire site. As with any artifact typology, a category may contain only one figure, but the patterning is more reliable when the sample size is larger.

Determining the Chronological Age of Petroglyphs Three relatively new and experimental dating techniques developed by Ronald Dorn—cation-ratio dating (CR), weathering rind organic dating (WRO), and varnish micro-lamination dating (VML)—have been used to obtain numerical age estimates for petroglyphs on the High Plains.34 Each of these techniques has strengths and drawbacks. Like most archaeological dating programs, however, they produce more reliable results when used in combination. Cation-ratio Dating CR dating is possible because of two properties of rock varnish: it contains both mobile and immobile elements (or cations), and trace elements such as potassium and calcium are leached from the varnish more quickly than elements like

40

L AWRENCE L. LOENDORF — THUNDER

AND

HERDS

TABLE 1.3. CHANGES IN QUADRUPED ATTRIBUTES THROUGH TIME Late Archaic Period Quadrupeds: 3000–1850 B.P. (A.D. 100) ƒ Quadrupeds may have horns or ears. ƒ If present, branching head appendages are never elaborate. ƒ Legs terminate without hooves. ƒ Quadrupeds and anthropomorphs never occur in the same rock art panel. ƒ Quadrupeds are not associated with spears or nets. ƒ Quadrupeds are usually associated with abstract forms such as circles, undulating lines, and meandering forms. Early to Middle Developmental Period Quadrupeds: 1850–1150 B.P. (A.D. 100–800) ƒ Quadrupeds have horns and branching antlers. ƒ If branching, head appendages are never elaborate. ƒ Legs sometimes terminate in inverted U-shaped hooves. ƒ Quadrupeds and anthropomorphs never occur in the same rock art panel. ƒ Quadrupeds may be associated with nets but not with spears. Late Developmental Period Quadrupeds: 1150–900 B.P. (A.D. 800–1050) ƒ Quadrupeds have horns and branching antlers. ƒ Some quadrupeds have elaborate, atypical antlers or other kinds of head appendages. ƒ Legs either lack hooves or occasionally terminate in round, ball-like feet. ƒ Quadrupeds sometimes occur with small numbers of anthropomorphs. ƒ Anthropomorphs sometimes occur in static postures near quadrupeds, but do not appear to be chasing them. ƒ Quadrupeds are usually associated with nets and arrows or spears. Diversification Period Quadrupeds: 900–500 B.P. (A.D. 1050–1450) ƒ Quadrupeds have horns and branching antlers. ƒ Some quadrupeds have elaborate antlers or other kinds of head appendages. ƒ Legs terminate without hooves. ƒ Stick-like anthropomorphs occur with quadrupeds and appear to be chasing them. ƒ Quadrupeds are always associated with spears and nets.

titanium. At any point in time, varnish contains different ratios of mobile and immobile elements, and by dating different rock surfaces at multiple points in time, a cation-ratio curve or timeline can be developed. Once the curve is established, it is a relatively simple process to sample and determine the cation-ratio of a petroglyph’s rock varnish and then place the image on the timeline.

1 — FOREGROUND

AND

BACKGROUND

41

A more difficult step in the process is determining the age of the rock surfaces upon which the curve is based. This is accomplished by obtaining AMS 14C dates for the organic matter, usually microcolonial fungi, living on the weathering rind of the rock and subsequently encapsulated by the rock coating. Taking pin-head-size samples from a petroglyph’s varnish, an analyst removes the organic matter for WRO radiocarbon dating. Dates obtained using this technique were initially thought to be as accurate as radiocarbon ages, but potential sources of error have been discovered since the technique was developed.35 Dorn originally collected the organic materials as a “bulk” sample and assumed that they were contemporaneous. He later learned that some rocks had been subjected to the development of organic matter on their surfaces at more than one time in the past. Perhaps most importantly, Dorn realized that the studies he had conducted on relatively recent basalt flows of known age “provided a very poor test, because they do not have a history of organic weathering before exposure. Basalt flows undergo organic weathering only after they erupt.”36 As rock surfaces age to the point that they develop a complete varnish, apparently some undergo erosion that destroys their protective layer. During this process, some organic matter from an older surface might become incorporated in a more recent surface, and the subsequent collection of organic material in a bulk sample would capture substances of different ages and therefore offer a false radiocarbon age. Dorn has continued to work with dating petroglyphs, and during this reevaluation, the use of VML dating of petroglyphs has gained momentum. VML dating is a correlative process in which the micro-stratigraphy of a petroglyph surface is compared with a previously defined time period. Underlying the technique is the fact that as rock varnish develops, the materials in its layers reflect regional climatic states. Changes from manganese-rich, cooler wet climates to warm dry climates with less manganese are recognizable in the varnish. Although Dorn originally investigated this technique for dating petroglyphs, Tanzhuo Liu has more recently studied thousands of rock varnish samples to establish a VML dating system spanning the period from the late Pleistocene through the Holocene.37 One important outcome of VML dating is that when it is used in combination with WRO and CR dates, confidence in the dating program increases. What effect has the preceding research had on the dating of petroglyphs in the High Plains region? First, it is important to recognize that most of the cationratio samples came from basalt dikes in southeastern Colorado. These dikes are millions of years old, but they have been buried for much of that time. Once basalt dikes are exposed, they tend to collapse into thick bands of rubble within a few hundred thousand years. Since the dikes of southeastern Colorado are still standing in well-formed walls, they are relatively young in geologic terms, and their varnish has not yet been subjected to multiple episodes of erosion, making them good candidates for petroglyph dating.

42

L AWRENCE L. LOENDORF — THUNDER

AND

HERDS

Another consideration is that when Dorn established the cation-ratio curve for the High Plains, he collected samples from at least 15 different localities to ensure that he would have enough usable organic matter.38 This geographically extensive collection strategy was replaced in Dorn’s subsequent research with a process in which organic matter in only a single sample, usually from a petroglyph, was dated and used as a point in the CR curve. Despite the problems with the use of the cation-ratio technique on the High Plains, the results have been promising. For example, a petroglyph of a figure holding a bow was dated by cation-ratio technique to the time period following the first use of the bow and arrow by prehistoric hunters in the region. In another compelling example, after cation-ratio dates were obtained for the upper portion of an abstract figure, archaeologists excavated the deposits covering the lower part of the figure and recovered charcoal that produced a slightly later radiocarbon date.39 Since soil deposition would have occurred after the petroglyph was created, radiocarbon analysis confirmed the correctness of the slightly earlier date produced by the cation-ratio method. Three VML dates have been obtained for sites discussed in this book. The dates are discussed in the next chapter, but I want to point out here that the VML ages confirm those produced by the cation-ratio technique. These tests provide considerable support for the accuracy of the CR technique, and they play a significant part in establishing the rock art chronology for the region. I present the dates for petroglyphs discussed in this book as relative ages, however, because I believe they are dependable when used in combination with other dating methods. Radiocarbon Dating An important part of my research has involved the radiocarbon dating of cultural deposits. Sometimes these are near to or associated with rock art sites or, in several instances including those described previously, samples were taken from datable material in the soil covering rock art panels. Of course, to the extent that these dates are more recent than the rock art, they are also minimum ages, but as shown in the example of a radiocarbon date on charcoal from deposits overlying a petroglyph, they can be very useful in setting chronological boundaries.

Developing Reliable Temporal Control From a methodological standpoint, I agree with rock art researchers Christopher Chippindale and Paul Taçon, who, when thinking about dating rock art, have discussed the utility of ideas expressed in Alison Wylie’s essay on archaeological reasoning.40 Wylie, a philosopher of science with a particular interest in archaeology, has observed that archaeological inference is often traditionally based upon “chains of evidence” built by adding one link to another. The problem with

1 — FOREGROUND

AND

BACKGROUND

43

this approach is that any chain is only as strong as its weakest link, and if one of the inferences in the chain is demonstrated to be false, then the argument so constructed falls apart. If, on the other hand, inferential arguments are constructed using a number of different types of evidence that taken together create an intellectual “cable,” and if the strands of evidence combine to form a strong cable, it will not be broken when one thread is later discovered to be weak. In subsequent chapters, readers will discover that I have approached the dating of rock art on the central High Plains by using the “cable” model. I have combined every source of evidence known to me to establish a cultural chronology, not as an end in itself, but as a prerequisite to understanding the patterns of change and development in the rock art record that temporal control makes visible.

NOTES 1 Although it was written 25 years ago, Donald Trimble’s Geologic Story of the Great Plains

remains an excellent source of information on the physiography of the High Plains. It is available through several sources on the Internet. 2 In 1821, a member of the Lewis-Dawson expedition was killed by a grizzly bear in the Purgatoire River valley (Friedman 1985:41). 3 The total number of 300 rock art sites is based on the Colorado and New Mexico archaeological site databases. The number is very conservative, since we know that Nancy Robertson, of Raton, New Mexico, has assigned temporary numbers to dozens of sites not included in the New Mexico database. 4 Trimble 1980:43–44. 5 Penn and Lindsey 1996. 6 The National Weather Service data presented in this paragraph are available at the following website: http://www.crh.noaa.gov/pub/?n=/ltg/cg_county_co.php. 7 Cassells 1992, 1997; Gilpin 2001; Ooten 1992; Robertson and Robertson 1975; Winter 1988; Zier 1999a:25–42. Dennis Gilpin describes prior research at Philmont Scout ranch; Robertson and Robertson describe rock art sites in the Raton area; and Joseph Winter presents an overview of the prior archaeological work in the Dry Cimarron River Valley, with a section by Nancy Robertson on the rock art of northern New Mexico. The other references provide good overviews of the history of avocational and professional archaeological research in Colorado. 8 Renaud 1936. Renaud mentions rock art in most of his reports, which are available on the Web through Special Collections at the University of Denver’s Penrose Library. 9 Lintz 1999. 10 An index to Southwestern Lore, the journal of the Colorado Archaeological Society, is available at the website: http://www.fortnet.org/casncc/SWLIndex/SWLArticles.htm#L; Tatum 1944. 11 Campbell 1969, 1976. 12 Burton 1971; Buckles 1964, 1971. 13 Buckles 1974, 1989. 14 Halasi et al. 1981. 15 McGlone et al. 1994, 1999; Dorn et al. 1990. 16 Winter 1988. 17 P. Schaafsma 1979; Wellmann 1978.

44

L AWRENCE L. LOENDORF — THUNDER

AND

HERDS

18 Loendorf et al. 1988. 19 Cole 1985; Loendorf et al. 1988; Loendorf 1989; Loendorf and Kuehn 1991. 20 Reed and Horn 1995. 21 Faris 1995; Keyser and Mitchell 2000; Mitchell 2004b. 22 Hartley and Vawser 2003. 23 Mitchell 2004b; Wintcher 2005. 24 Kalasz et al. 1999; Gilmore et al. 1999; Winter 1988. 25 Stephen Kalasz et al. (1999:55) recommend that rock art be considered a complementary data

set to other archaeological remains in the Arkansas River basin.

26 An expanded discussion of rock art site definitions appears in Loendorf et al. (1988) and

Loendorf (2001).

27 Julie Francis (2001) provides a good overview of style and classification in rock art studies. 28 Loendorf (1989:75–118) describes the criteria on which the classification is based. The Piñon

Canyon boulder sites are described in more detail in Chapter 2.

29 Greer 1995; King (1978) and Whitley (1982) use a three-pole regression type of seriation. 30 In the initial typology, type B was combined with type A. The typology and seriation categories

are explained in greater detail in Loendorf (1989:75–118; 338–40) and Loendorf and Kuehn (1991:262–65). 31 Dates reported by Heizer and Baumhoff are now recognized as inaccurate, with the result that curvilinear abstract figures are not necessarily older than rectilinear ones, as reported in their Great Basin study. See Whitley (2000) for a discussion of the revised dating of Great Basin petroglyphs and Hedges (1982) for a discussion of abstract forms. 32 In his discussion of Great Basin abstract petroglyphs, Hedges (1982) also concludes that curvilinear abstract figures are not necessarily older than rectilinear ones. 33 Tratebas (1993) employs a similar dating approach based on attributes that are added to or dropped from a petroglyph tradition. 34 See Dorn (2001:172–83) and Whitley and Loendorf (2005:925–28) for more discussion of these dating techniques. 35 Dorn 2001:179–83. 36 Dorn 1996:10. 37 See http://www.vmldatinglab.com/ for more information about this dating technique. 38 Dorn 1989:134–35. 39 Loendorf 1991. 40 Taçon and Chippindale 1998:92–93; Wylie 1989.

2 ROCK ART OF THE ARCHAIC PERIOD

I NTRODUCTION n the 150 years that have elapsed since systematic archaeological research was first undertaken, a paramount concern has been to place the artifacts and features created by past human activities into their proper chronological sequence. The development of an accurate temporal progression of archaeological materials and their contexts, made possible by careful comparative analysis and increasingly sophisticated dating techniques, has brought order to a once-confusing panorama of individual sites and their contents. The study of rock art has benefited from archaeology’s systematizing efforts, particularly its transformation of archaeological remains from undifferentiated eons on the High Plains—beginning with the human migration into North America and ending with the present—into discrete periods marking changes in human lifeways. As a result, two adjectives are routinely applied to the petroglyphs found at the three sites discussed in this chapter. The word archaic identifies their placement within a 5500-year period on a timeline that begins around 7500 years ago and continues forward until 2000 B.P., and the word abstract summarizes their predominant stylistic properties. In this chapter I will first describe the individual histories and rock art inventories occurring at the Clay Creek, Glorieta Mesa, and Piñon Canyon boulder sites and then will provide an overview and comparison of the similarities and differences between these selected groups of petroglyphs that are representative of the Archaic Period on the southern High Plains.

I

45

46

L AWRENCE L. LOENDORF — THUNDER

AND

HERDS

THE CLAY CREEK PETROGLYPHS On the southern High Plains, the first five months of 1965 had been significantly drier than normal. Because the meteorological Big Picture only becomes clear in retrospect—once weather data are collected and analyzed—residents in the five-state area did not yet know that they were in the second year of a three-year drought. By June, instead of having received the average 5.45 inches of precipitation that usually occurred between January and May, Lamar, Colorado, had recorded a mere 1.98 inches. Farmers needed rain so badly that almost any approach to drought relief was considered reasonable. Twenty-eight miles to the east in western Kansas, clouds were seeded, and some church groups held special services at which members prayed for rain. The storm clouds that crossed the Rocky Mountains and moved in from the west on Sunday evening, June 13, and again on Tuesday morning, brought heavy local showers. With them came the promise of productive yields in the farmlands bordering both sides of the Arkansas River as it flowed east from Lamar, through the towns of Granada and Holly, to the Kansas border and beyond. Unfortunately, as local newspapers reported, floods of devastating proportions occurred in the next two days, as silt, debris, and river water inundated the region, creating a disaster area in the river valleys along the front range of the Rocky Mountains and out onto the plains. The National Guard was mobilized to help with rescue efforts, and President Lyndon Johnson declared eastern areas of the state eligible for federal relief.1 Statewide the death toll rose into the double digits, residents were forced to evacuate, and dozens of houses were lifted off their foundations and deposited a mile away. Persons marooned on their roofs awaited rescue by helicopter. The total amount of rainfall varied from place to place, but the 14 inches recorded south of Lamar set a one-day record that has yet to be broken. The magnitude of the inundation becomes apparent when it is compared with the town’s average annual rainfall of 15.20 inches, a statistic compiled from records for the 87year period between 1918 and 2005.2 Twenty-five years later, James W. Casteet, the former postmaster of Holly, expressed the sentiments of many survivors when he said, “I do hope, in the future, when the churches pray for rain, they will specify the amount and over what period of time it will fall.”3 Often archaeologists are among the few beneficiaries of catastrophic events such as those occurring in the Lamar-Holly area in June 1965. When used in an archaeological context, the word dig is a reminder that remnants of the prehistoric past are most often buried in sediments whose removal is laborious, time-consuming, and expensive. There are exceptions to this generalization, however, and once the floodwaters had receded after the extraordinary deluge of 1965, it was discovered that vast amounts of surface sediments had been washed away, here and there exposing geological outcrops that had lain hidden for centuries under layers of alluvium.

2 — ROCK ART

OF THE

ARCHAIC PERIOD

47

One of these formations, known as Dakota sandstone, is the primary exposed sedimentary formation across the region. The extent of its deposits near the Clay Creek dam are unknown, but the post-flood discovery of a previously buried rock art site by Bureau of Reclamation hydrologist P. O. Abbot immediately established that at least one sandstone surface had been available for use by prehistoric huntergatherers. Although not an archaeologist, Mr. Abbott recognized the importance of the newly visible panel of rock art images and brought it to the attention of William Buckles, a professor of archaeology at the University of Southern Colorado.4 As I noted in my brief historical review of regional rock art research in chapter 1, Buckles was an experienced field archaeologist who had recorded and analyzed rock art at several High Plains sites. Buckles realized that the deeply pecked, wide-line, abstract petroglyphs at Clay Creek were similar in motif and technique to designs that he and other archaeologists had identified as the oldest rock art in western North America (Figure 2.1). He also noted that part of the large panel was still covered by a soil type called Piney Creek alluvium, an identification later verified by Glen Scott, a U.S.G.S. soils expert, who confirmed that the well-stratified Piney Creek sands, silts, and clays with inter-layered gravels were middle Holocene in age. Since Piney Creek sediments have been dated between 1850 and 3500 years of age, Buckles was confident that the rock art panel had been created within this

FIGURE 2.1. Close-up photograph of exposed Clay Creek petroglyphs in the Dakota sandstone formation. Note the carbonate accretions adhering to the panel, which was exposed by the 1965 flood. Photograph by Lawrence Loendorf.

48

L AWRENCE L. LOENDORF — THUNDER

AND

HERDS

time frame.5 Attempting to determine more specifically when the petroglyphs had been buried—which would provide an “at-least-as-old-as” temporal parameter—he searched for a suitable organic sample in the soil still covering parts of the panel to use for radiocarbon dating. After two years, he found pieces of charcoal associated with the remnants of a hearth and obtained an age of A.D. 100 ± 100 on the samples.6 This date is important because it establishes that the underlying petroglyphs were produced no more recently than 1900 years ago, although for reasons to be discussed shortly, they may be at least one thousand years older than the soils that covered them.

Abstract Forms at Clay Creek When the word abstract is applied to visual objects, it usually means that the forms of those objects do not have the identifying characteristics of entities recognizable in the natural world. Although abstract forms can be linear, circular, rectangular, or some combination of all these shapes, they are not representative of the human body or plants or kitchen tables, or any of the things in the threedimensional world that can be observed and named. When the petroglyphs at Clay Creek are described as abstract forms, it means that they don’t resemble the persons or species with which their creators would have coexisted in daily life. As a group, the figures are more circular than rectangular and consist predominantly of detached, connected, or bisected circles. Approximately one-third of the figures are independent of one another, but the remaining two-thirds are either connected by long undulating lines or are attached to form a larger abstract component within a long, horizontal group of figures (Figure 2.2). The inventory of shapes includes concentric circles, circles with either tails or lines resembling rays extending from the perimeter, and circles divided by radial lines into pie-shaped wedges like wagon wheels. A recurring motif is an oval whose interior is divided horizontally into five or six units and bisected vertically by a line that usually extends beyond both ends of the oval. These layered, lollipop-like figures are found across the site in several sizes, ranging from 15 to 75 centimeters in height. Other motifs have been classified as concentric circles, but a better term might be double-outlined circles, since the inner circle is inscribed close to the outer one, making the figure look like a bicycle tire or a steering wheel. Some of these circles are bisected by a vertical line that extends beyond the perimeter at each end, while another group has a similar line that divides the circles horizontally. Many of the figures are more difficult to describe. They tend to have long, undulating, curving, or wavy lines that occasionally loop around to form enclosed circular areas which in some instances are connected to each other. They appear randomly distributed and can cover several meters of a sandstone outcrop. Not only are there no anthropomorphic figures, quadrupeds, or other recognizable forms at the Clay Creek site, there are no images that appear to be attempts to produce stylized

across the horizontal axis is approximately 4 meters. Illustration by Davíd Joaquín from a tracing by William Buckles.

FIGURE 2.2. The middle section of the Clay Creek petroglyph panel representing about 85 percent of the total number of petroglyphs. The distance

50

L AWRENCE L. LOENDORF — THUNDER

AND

HERDS

or abstract versions of real-world entities. The absence of figurative designs is significant, since it is likely that if such figures were being produced prior to 2000 years ago, when the site was buried, they would have been included in the inventory at this large site. There are also no cross-hatch designs, diamonds or diamond chains, equilateral crosses or vertical lines with multiple crossing lines, or dots arranged in groups or patterns. All of these motifs are common at other Archaic Period petroglyph sites on the High Plains, so their absence at the Clay Creek site may indicate that the rock art panels there predate the introduction of such patterns. It is common at rock art sites containing the meandering forms just described to find areas or places where one image has been inscribed over another, completely different image. Although Buckles inspected the sandstone surface carefully, looking for examples of this kind of superimposition, none were found. His search did reveal evidence that some of the figures had been repeatedly repecked at a later time. The lines, in some cases 2.5 centimeters in depth, conformed to and made deeper the original outline of the reworked petroglyphs.7 Questions about when the reworking occurred are to some extent answered by radiocarbon dating of the overlying soils that buried the site and have eliminated the possibility that the repecking took place in the succeeding Late Prehistoric Period. All return visits to the prehistoric site must have taken place between the time the petroglyphs were made and when they were buried 1900 years ago. It has been occasionally suggested that rock art represents spontaneous acts of expression by the prehistoric equivalent of contemporary graffiti artists, but the petroglyphs at the Clay Creek site negate this simplistic assertion. The fact that one or more figures were inscribed and, at a later time, the motifs were made more prominent by the same or another person is strong support for the argument that rock art was an integral part of the cultural traditions of the Archaic Period. Of equal significance is the fact that some of the Clay Creek figures recur frequently at other rock art sites on the High Plains, which suggests that the motifs had meaning that crossed cultural boundaries. Whether the abstract forms were created in conjunction with trance-related activities, as is frequently suggested, or were used to mark territorial boundaries, or can be accounted for by some unknown reason is a question that continues to be debated by rock art scholars worldwide. There is much to be learned, however, from a study of the distribution of Abstract Period forms across the region, and the lack of consensus about their purpose in no way reduces the importance and information potential of the Clay Creek site and others like it.

THE PIÑON CANYON B OULDER SITES The power of water to transform landscapes has impacted rock art in more than one way. Not only has it created conditions leading to the discovery of sites like

2 — ROCK ART

OF THE

ARCHAIC PERIOD

51

the Clay Creek petroglyphs, it has made accessible the surfaces later used by humans as the raw material on which they incised or painted graphic designs. The story of the climatic events that had a defining effect on the landforms— and therefore the rock art—at Piñon Canyon is complicated. It begins about 10,000 years ago at the end of the last ice age, as temperatures gradually increased and the ice packs in the glaciated peaks and high valleys of the Rocky Mountain range in Colorado began to melt. The process was not continuous, however. Conditions reversed several times and the glaciers again advanced, most recently during the 500-year period called the Little Ice Age that began in the fourteenth century.8 Although the mercury in the global thermostat appears to be rising once again, it would be a mistake to think of glaciers as extinct. In a recent survey of the American West, scientists at Portland State University report that although in Colorado today most perennial ice patches are “tattered remnants” of their former early Holocene grandeur, 14 small glaciers, primarily in the Front Range, still remain.9 The path taken by the often high-velocity glacial meltwater that surged periodically from the eastern slopes of the Rockies was determined by the fact that the mountains lie west of and at a higher altitude than the plains of eastern Colorado. The force of the rushing water followed naturally occurring undulations in the prairie surface and, over time, cut several river channels and numerous southeasttrending arroyos through sediments and bedrock. Unlike rivers, whose currents are constant though often variable, arroyos are today characterized by intermittent water flow during the spring melt and the summer rainy season. The arroyos at Piñon Canyon, while relatively shallow and dirt-sided along their upper reaches, are often transformed into twisting, narrow sandstone canyons by the time they join the Purgatoire River to the east. Another factor affecting the future placement of petroglyphs at Piñon Canyon originated in the even more remote past and resulted from volcanic activity in the area 14 million years ago. Tectonic forces deep within the earth’s crust propelled basalt-laden magma upward, fracturing the existing bedrock and creating a series of fissures that were filled with molten rock. These underground features were originally covered with sediments that concealed their subsurface structure, but they were gradually exposed by erosion, revealing spectacular basalt formations resembling the backbones of ancient creatures. At some point in the past, all along the northern side of the Hogback, rapidly flowing glacial meltwater carved out the channel later given the name Van Bremer Arroyo. At times of peak flow, currents scoured the northern slope of the Hogback, transporting watermelon-size chunks of basalt downstream.10 These boulders were deposited along both banks of the arroyo where, after more time passed, native peoples inscribed hundreds of petroglyphs on their surfaces at a half-dozen localities collectively referred to as “the boulder sites” (Figure 2.3).

52

L AWRENCE L. LOENDORF — THUNDER

AND

HERDS

FIGURE 2.3. A view of 5LA5598, one of the Piñon Canyon boulder sites. The petroglyphs occur on exposed basalt boulders that are, on average, about the size of a watermelon. Photograph by Peter Halter from the collection at the United States Army, Fort Carson Curation Facility.

Rock Art at the Boulder Sites The dark-colored, non-lithic material called varnish, patina, or crust that accumulates on the surface of rocks over time is well developed on the basalt boulders in Van Bremer Arroyo and makes them ideal surfaces for inscribing petroglyphs. Varnish is composed of microscopic forms of familiar chemical elements such as helium, carbon, and aluminum, as well as airborne organic substances, and the act of pecking or incising a design removes these accretions and exposes the inner, light-colored stone. Unlike the rock art panels at Clay Creek, which had been covered with sediments prior to their exposure, the boulder sites in Van Bremer Arroyo remained uncovered from the time the first petroglyphs were placed there. At other rock art sites in the American Southwest, such extended visibility often results in heavy reworking of the original images. Petroglyph panels at Blue Bull Cave in Canyon de Chelly, for example, have up to seven different layers of superimposed images, with the oldest dated to about 2000 years before the present. At the Piñon Canyon boulder sites, however, there is little evidence that related or wholly different images were added to the mix of designs inscribed during the Archaic Period.

2 — ROCK ART

OF THE

ARCHAIC PERIOD

53

The overwhelming majority of the petroglyphs at the boulder sites are abstract figures, but unlike contemporaneous panels at Clay Creek and Glorieta Mesa (described later), the design inventory includes quadrupeds and a few other representational figures. A closer look also reveals that the abstract forms differ in significant ways from similar images at the two other sites (Figure 2.4). For instance, the layered, lollipop figures found at Clay Creek also occur at the boulder sites, but here they are more rectangular than oval and are frequently augmented by small circles either adjacent or attached to the main figure. Like the oval lollipop figures at Clay Creek, consistencies in the form of the rectangular examples at the boulder sites demonstrate that they are not simply random variations on a typical theme.

FIGURE 2.4. Selected petroglyph boulders from the boulder sites at the Piñon Canyon Maneuver Site. Their size is variable, but as a group they average about 40 centimeters across. Scale drawings by Janet Lever, Linda Olson, and Anne Hayes, produced in 1988 as part of a contract obligation with the United States Army, Fort Carson.

54

L AWRENCE L. LOENDORF — THUNDER

AND

HERDS

Concentric circles also occur at the boulder sites, but rather than exhibiting any uniformity of design, their interior horizontal and vertical lines appear haphazardly placed. The inner circle is also more centrally placed—like a bull’s eye—rather than close to the outer circle, like the bicycle tire examples at Clay Creek. When the inventory of circular motifs is considered as a whole, there are fewer embellished circles at the boulder sites than at either Clay Creek or Glorieta Mesa. The distinguishing characteristic of the petroglyphs at the boulder sites is, however, the presence of quadrupeds. These figures have the generic features of their biological counterparts—head, body, legs, and tail—but are small, solidly pecked, and stiff-legged (Figure 2.5). Their body shape can vary from rectangular to boat-shaped, but their tails are always straight appendages sticking out horizontally from the back end of their bodies. Heads are rounded and have horns or ears,

FIGURE 2.5. Quadruped petroglyphs from the Piñon Canyon boulder sites. Note how some of the figures emerge from the abstract designs, while the animal figures in a circular format are independent. The boulder with the circular arrangement of quadrupeds is about 50 centimeters across. Scale drawings by Janet Lever, Linda Olson, and Anne Hayes, produced in 1988 for the United States Army, Fort Carson.

2 — ROCK ART

OF THE

ARCHAIC PERIOD

55

but none have branching antlers. The legs of the most carefully formed, boatshaped figures end in U-shaped feet, but the other quadrupeds are depicted without hooves or feet of any kind

Pits and Petroglyphs Two interesting pit features were discovered among the 264 boulders at the largest of the Piñon Canyon boulder sites. Surrounded by petroglyph-bearing boulders and clearly associated with them, the two slightly oblong, funnel-shaped pits had surface openings measuring 1.5 by 2 meters and 2 by 2.5 meters.11 At the time of their discovery the pits were 40 centimeters deep, but archaeological excavation revealed that the deposits consisted of sediments that had accumulated through time and that their original depth had been 70 to 80 centimeters (Figure 2.6). The pits’ meager contents provided no clues about how they were used in the past. One pit contained so little material that excavation was abandoned at a depth of 30 centimeters. The other pit contained a small arrow point on the surface of the deposit, and one antelope phalange, small amounts of ash with charcoal flecks, a few FIGURE 2.6. A pit feature at 5LA5598, one of the chipped stone flakes, and severPiñon Canyon boulder sites. The pit measures 2 by al fairly sizable pieces of red 2.5 meters across. Charcoal from the base of the pit ocher. While these remains was dated at A.D. 100 ± 100. Photograph by Lawrence were not concentrated, most Loendorf. were recovered between 18 and 30 centimeters below the surface. Excavation continued to 40 centimeters and revealed that the pit bottom was constricted and filled with rocks. The arrow point found on the surface of the deposit could have been discarded at any time after the pit was abandoned, so attributes such as its style and raw material cannot be used to date the deposit. Although a radiocarbon date of 391 B.C. was obtained on a piece of charcoal at a depth of 30 centimeters, the question of whether the charcoal and the petroglyphs surrounding the pit are contemporaneous is currently unanswerable.12 Identification of the context in which the two pits functioned in the past is based on what the medical profession refers to as a “diagnosis by exclusion.” When confronted by a set of ambiguous symptoms that do not immediately suggest a readily identifiable disease, doctors will compare the symptoms with a medical template that enables them to say, “Well, it’s not tuberculosis” or “it’s not appendicitis.”

56

L AWRENCE L. LOENDORF — THUNDER

AND

HERDS

By ruling out the diseases that can’t be causing the symptoms, clinicians can often make better decisions about which laboratory tests might produce direct evidence about the cause of the medical problem. This same approach is sometimes applied to ambiguous archaeological features. For instance, it is well known that the remains of prehistoric pit houses usually include post holes, packed floors, prepared hearths, or some combination of these features. Since the pits at the boulder site contained none of these properties, archaeologists are on firm ground when they conclude that the pits are not the remains of houses. Similarly, the absence of heat-cracked stones and fire-reddened or charred side walls also indicates that the pits were not used for roasting root vegetables or other plants, a conclusion supported by the absence of metates or other tools associated with plant processing. Another possibility is that the pits may have been used for storage, but since they do not have any properties in common with a known storage pit dug into the alluvium near Van Bremer Arroyo—which had straight sides and a cache of manos in the bottom—it is likely that the pit is related to some other activity. It has been suggested that as part of a strategy for catching eagles, perhaps prehistoric hunters hid in the pits, next to a tethered rabbit set out as bait. The problem with this interpretation is that known eagle-catching pits on the High Plains are invariably located on high, open ridges having a good western exposure. Because the boulder site pits do not meet this important topographic criterion, it is unlikely that they were used for catching eagles. Ethnological studies of hunting practices have prompted the suggestion that the pits served as “calling stations” for shamans, who would hide there, meditating and praying, unobserved by the antelope that they hoped to attract into Van Bremer Arroyo, where hunters waited to ambush and kill them.13 Although support for this suggestion comes from the hunting-related rock art imagery at other sites along the arroyo, it is insufficient evidence for a conclusive diagnosis of the pits’ former use. From an archaeological perspective, the presence of pits at the boulder site underscores both a general principle and a specific research imperative. First, it is generally true that there is often much to learn from relating rock art to its archaeological context; and, second, other High Plains Archaic rock art sites should be examined to determine whether similar features are present and related to the rock art inventory.

Archaic Rock Art at Glorieta Mesa Glorieta Mesa is a New Mexico landform with a dramatic human past. Today its almost 200 square-mile surface is peaceful, presided over by birds, small mammals, and the occasional mountain lion. These species share the grassy meadows and piñon-juniper woodlands with grazing cattle belonging to the handful of local

2 — ROCK ART

OF THE

ARCHAIC PERIOD

57

ranchers who maintain the western tradition of livestock raising. Most of the steepsided sandstone and limestone mesa, whose elevation exceeds 8000 feet, is under the management of the U.S. Forest Service, which controls use of the irreplaceable natural and cultural resources needing protection from human encroachment and development. In historic times, the serenity of the mesa was disturbed first by a trickle and then by a stream of eastern adventurers and settlers pursuing the opportunity to establish themselves in the so-called unoccupied lands of the West. Prior to the sixteenth century, an aboriginal footpath followed the northern curvature of the mesa and threaded its way through a natural break or canyon—now known as Glorieta Pass—in the Sangre de Cristo Mountains. By the mid-nineteenth century, the path had become part of a major thoroughfare whose name—the Santa Fe Trail—is so ingrained in the public imagination that it still evokes the sights and sounds of wagon trains, immigrant families, cattle rustlers, and gunsmoke. In the last week of March 1862, the cacophony of war unexpectedly enveloped Glorieta Pass as events connected with a turning point in the American Civil War, often referred to as “the Gettysburg of the West,” took their toll.14 The conflict on New Mexican soil, far from the capitals and most highly contested terrain fought over by Union and Confederate troops, involved the head-on collision of troops from both sides. Many casualties occurred in the next two days—the final count was more than 280 soldiers—but a surprise Union raid on the Confederate supply depot was devastating to the morale of the southern troops. Realizing that they had been outwitted and deprived of all essential material support, the troops had no choice but to abandon their goal of territorial expansion, acknowledge defeat, and begin their long, dispirited withdrawal from New Mexico. With their departure, life on Glorieta Mesa was once again peaceful, although not unchanging. In the succeeding century and a half, the AT&SF railroad came through Glorieta Pass; New Mexico achieved statehood; and many small Hispanic communities evolved into prosperous towns and, in a few instances, cities. Interstate 25 now wends its way east from the state capital in Santa Fe, skirts the northern end of Glorieta Mesa, and then turns north toward Denver and beyond. Only a handful of the thousands of contemporary residents who live and work in the ever-expanding greater Santa Fe area are aware that several thousand years before commuter and interstate traffic began to flow through Glorieta Pass, and millennia prior to the clash of Civil War armies there, mobile, indigenous families were present and left their mark on Glorieta Mesa. According to Forest Service archaeologist Brent Abel, in the late 1800s a ranching family in the area named Huddlestone had noted the presence of a petroglyph site on the mesa that many years later—in the late 1970s—Abel’s agency briefly described and designated as site 6 in the Forest Service archaeological inventory.15 In the late 1980s, Abel himself went looking for the sandstone surface whose rock art had survived

58

L AWRENCE L. LOENDORF — THUNDER

AND

HERDS

everything that nature could inflict on their surfaces, as well as the hooves of deer, elk, and cattle, the boots of Union soldiers, and the wheels of dirt bikes and other off-road vehicles. He not only found site 6, a massive surface covering more than 100 square meters, but also another, equally large but partially buried surface that has been designated site 147.16 Today, as in archaic times, the horizontal sandstone rock faces, flush with the surrounding terrain, are covered with hundreds of engraved images that stare skyward (Figure 2.7). Abel realized immediately the importance of protecting and recording the petroglyphs at both sites, so he turned to two sources for help with such an extensive project. The American Rock Art Research Association (ARARA), established 40 years ago by professional and self-taught rock art researchers to support the recording and preservation of rock art resources, was asked to provide guidance and expertise. The Forest Service’s Passports in Time (PIT) program, which offers the public the opportunity to volunteer for short-term fieldwork assignments with scientists from many disciplines, was approached for help in finding persons interested in being trained to record the rock art using a variety of techniques.17 Things came together in the summer of 1992, when ARARA researchers Frank and A. J. Bock and a group of PIT volunteers arrived on Glorieta Mesa and began the initial recording effort.18 They took photographs, made scale drawings, and drew detailed maps showing the distribution of the petroglyphs. At both sites,

FIGURE 2.7. Petroglyphs on the exposed sandstone surface at Glorieta Mesa. Photograph by Lawrence Loendorf.

2 — ROCK ART

OF THE

ARCHAIC PERIOD

59

sections of the rock art panels were covered with sandy soil, so crews used whisk brooms to sweep them clean. As they did so, they noticed that at the edge of one grouping at site 147 the petroglyphs disappeared under a layer of soil that was between 25 and 30 centimeters deep. Burial presented the possibility that the overlying sediments contained archaeological materials, particularly charcoal or other organic substances, that could be radiocarbon dated to establish a minimum (“at least as old as”) age for the rock art below. Because excavation is not undertaken spontaneously by responsible archaeologists, but requires the development of a research design and specific recovery goals, discovery of what the soils contained and the nature of the rock art underneath had to be deferred until the next summer season. In 1993 and 1994, teams returned to Glorieta Mesa, outfitted with stakes, strings, shovels, trowels, and screens to begin the labor-intensive task of removing the overburden concealing the petroglyphs at both sites 6 and 147 (Figure 2.8).19 Work included analysis of the sediments by John C. Phillips, a Forest Service soil specialist.20 Ronald Dorn, the archaeological dating specialist whose work with cation-ratio analysis was described in greater detail in Chapter 1, took samples of the varnish that had developed over time in the petroglyphs.21 Not many artifacts were recovered during excavation. The sediments contained a half-dozen possible grinding tools called manos and one chipped stone flake. A projectile point found on the surface of one deposit was photographed, but it was not collected. There is useful information in the photographs, however, which show a stubby, deeply notched point whose stem and tip were unfortunately broken in a

FIGURE 2.8. One of several excavations to expose buried petroglyphs at Glorieta Mesa. Note the depth of the soil covering the petroglyphs. Photograph courtesy of Brent Abel, United States Forest Service.

60

L AWRENCE L. LOENDORF — THUNDER

AND

HERDS

way that makes identification of the point’s overall configuration impossible. A metric scale next to the projectile point reveals that its width and length were 2.5 centimeters, although by mentally extending the sides to the point at which they would have converged and formed a tip, it is clear that the length would have been approximately 2.75 centimeters. These dimensions, coupled with the 1.4-centimeter-wide broken base, makes the point similar to other Early to Middle Archaic stemmed or notched points from the region.22 The age of the petroglyphs at site 147 has been established by a series of radiocarbon dates for soil samples overlying the site surface that reveals an interesting chronological correspondence between the petroglyphs and the projectile point. John Phillips’s soil study determined that the sediments covering the petroglyphs were well developed and that significant clay and calcium carbonate accumulations occurred at the bottom of the profile.23 In an attempt to date the soils, charcoal in the sediments covering the petroglyphs was submitted for radiocarbon analysis, which determined that the source was apparently a natural fire and not burning associated with a hearth on the petroglyph’s surface. Dating natural charcoal can be misleading because samples may have been transported from an older nearby surface. In this case, however, the charcoal on both sites dated consistently to more than 2500 years ago, and several samples dated to between 4000 and 4500 years of age. The most important dates occurred on two samples collected from site 6. One sample from 15 centimeters above the petroglyph panel had a date of 2570 ± 70 B.P., while a sample from 4 centimeters above the panel had a date of 4480 ± 70 B.P. Since the samples were stratified in a well-developed soil profile, these dates indicate that the petroglyphs are more than 5000 years old.24 Ronald Dorn calculated cation-ratio ages for the exposed petroglyphs at the sites, and these also ranged between 4000 and 5000 years old. All of these dates—radiocarbon ages, the age of the projectile point, the accumulated soils, and the cationratio dates—are strong confirmation that the petroglyphs were created in the Early Archaic or the beginning of the Middle Archaic Periods.

The Glorieta Mesa Petroglyphs A basic rule of rock art research is that an observer must refrain from projecting onto images from the past any identifications or meanings originating in the present, unless—and this is a very important condition—there is biological, ethnographic, historical, or some other kind of research supporting such an identification. It is true, however, that rock art researchers are carefully developing bodies of theory that in some settings allow them to make explanatory statements about the rock art images they study. At the simplest level, there are many instances in which figures of four-legged animals so closely illustrate the characteristics of a particular species—a deer, for instance—that a claim that the figure is a cervid is sustainable. But because a perfectly depicted image of a deer is still

2 — ROCK ART

OF THE

ARCHAIC PERIOD

61

a symbol, representing either a living animal or a concept known only to the cultural milieu in which the image was created, most researchers resist the impulse to make definitive pronouncements about what a rock art image means without considerable research. Such restraint is especially necessary when examining the 226 petroglyphs at site 147 and the 269 engravings at site 6, although for strictly descriptive purposes some comparisons with forms occurring in nature will be made here. In general, it can be said that the petroglyphs at sites 6 and 147 share many of the properties of abstract figures at other High Plains Archaic rock art sites. There are many squiggly, meandering lines and a variety of circular and rectangular motifs, but here these basic elements are combined and juxtaposed in more complicated, provocative arrangements. For instance, at the Clay Creek site it is common to find tailed circles—round shapes from which a wavy line extends from one point of the circumference, like the string on a balloon. At Glorieta Mesa, the same basic form occurs, but the circle is less round and the tail is zigzagged or serpent-like. There are also several V-shaped figures intersected by a perpendicular straight line that closely resemble bird tracks. In at least one instance, such a “bird foot” is almost touching an egg-shaped image that is divided by five interior straight lines. Many circles are simply concentric, but they can also be embellished by rays projecting from their outer ring. In some instances circles are divided into segments by lines radiating from the center, which makes them look like another natural form, the sand dollar. Some concentric circles have smaller, irregularly shaped “petals” at intervals around the outermost ring. Nested semicircles can be unrelated to other forms, but sometimes they curve around, but do not touch, the edge of another circular or rectangular figure. Grid-like forms are also plentiful. Some are box-like, bisected by horizontal lines and sometimes joined along one edge, creating the impression of an open book having six lines of “type” on each “page.” There are several examples of vertical lines from which nine or more horizontal lines extend, like ribs radiating from a sternum. Starburst designs made up of as many as ten straight lines radiating from a central point are numerous. The inventory at site 147 includes one figure that closely resembles the track made by a bear, and another has much in common with an ungulate hoofprint. There is even an image that someone allowing their imagination full rein could argue was a human footprint. Whether these and other types of provocative figures are rare or common on Glorieta Mesa can only be answered by a systematic survey of this very large geographical feature. The time, money, and effort that would be necessary to double, triple, or quadruple the sample of mesa-top petroglyph sites—and archaeology’s knowledge about rock art variability at this very important location—would, however, return the investment a thousandfold.

62

L AWRENCE L. LOENDORF — THUNDER

AN OVERVIEW

OF

AND

HERDS

ARCHAIC ROCK ART

Despite the fit between the time period and the distinguishing characteristics of the petroglyphs I have discussed in earlier sections of this chapter—and the fact that Archaic petroglyphs are usually produced by a distinctive pecked technique—there are noticeable differences in their form and placement. This variety is not surprising, given their extensive geographic distribution and the thousands of years during which they were produced. For a period of five millennia, indigenous peoples living on the High Plains left dozens of major sites—and scores of localities with only a small number of pecked abstract petroglyphs—in eastern Colorado, New Mexico, and the panhandles of Texas and Oklahoma. In this chapter I have focused on Archaic abstract petroglyphs within the geographic boundaries of the central High Plains, but this distinctive rock art phenomenon has been recorded over a wide area of the American West. Linea Sundstrom has described several sites in the southern Black Hills of South Dakota which, in my experience, represent the northernmost extension of Archaic abstract petroglyphs. Polly Schaafsma has reported on extensive sites, over 1200 miles (1931 kilometers) to the south, in the Chihuahuan desert of southeastern New Mexico and northern Mexico.25 West of the High Plains, Archaic abstract petroglyphs are the predominant rock art form in the high interior basins and plateaus west of the Front Range in Colorado. They are found in Nevada’s Great Basin and continue into the Sierra Nevada Mountains of California. It is in the latter region that Robert Heizer and Martin Baumhoff initially identified differences between Archaic petroglyphs that they thought justified dividing the category into two styles, termed Abstract Curvilinear and Abstract Rectilinear.26 Just how far east beyond the panhandles of Oklahoma and Texas, which are almost 1000 miles due south of the Black Hills, abstract petroglyphs can be found is not yet known. Because abstract petroglyphs occur in all post-Archaic time periods, although more dispersed and in fewer numbers, it is often difficult to say with certainty that particular abstract petroglyphs are Archaic in age. On the Uncompahgre Plateau of southwestern Colorado, for example, abstract figures are spatially integrated with more realistic forms and may be contemporaneous with them.27 Research has determined that in much of Nevada, Numic-speaking groups, and perhaps others, produced abstract petroglyphs in post-Archaic time periods. Similarly, there is some overlap on the High Plains between Archaic abstract petroglyphs and those from later time periods, but at many sites abstract figures are unaccompanied by representational images and are likely therefore to date to the Archaic Period. Despite the extensive geographical distribution of Archaic abstract petroglyphs, the scholarly attention they have received has not been correspondingly intensive. This neglect may in part reflect a tendency to focus on rock art imagery that is at least superficially recognizable. Sites containing representations of human or animal figures are thought to provide researchers with a higher inter-

2 — ROCK ART

OF THE

ARCHAIC PERIOD

63

pretive baseline than those—such as Glorieta Mesa, Clay Creek, and Piñon Canyon—whose inventories are composed of squiggly lines and zigzag patterns. Even though the three preceding sites contain predominantly abstract figures, in many respects the images differ significantly from one another. Identifying and analyzing these differences is challenging work, and very few archaeologists have been willing to attempt so complex an inquiry. One exception is Charles Mobley, who, with Paul Steed, studied the rock art in the Los Esteros area of eastern New Mexico, near Santa Rosa.28 Mobley’s research was part of a larger archaeological inquiry designed to answer questions about the Los Esteros cultural sequence. Were the descendants of Archaic foragers, whose archaeological remains begin the sequence, responsible for the later Pueblo sites in the area? Or were the local Pueblo sites produced by visitors from western Pueblo villages who intermittently exploited the region’s resources? A third possibility was that local sites were created by indigenous hunters and foragers who exchanged subsistence items for Pueblo artifacts, some of which entered the archaeological record. Mobley hoped that analysis of the 17 rock art sites in the study area would illuminate some of the preceding research questions, perhaps by demonstrating temporal or spatial clustering, or affinities among types of figures. He created a rock art typology consisting of 43 mostly abstract motifs and then subjected the data to multivariate Q-mode statistics to learn if there were significant correlations among the motifs. When no pattern of association in the data emerged, Mobley concluded that the motifs were equally distributed across the sites. What Mobley apparently didn’t realize was that not all statistical techniques are applicable to particular archaeological data sets. Sample size is a major factor in the production of definitive patterning, and Mobley’s 17 sites included 6 sites with fewer than 10 rock art elements and an additional 7 sites with fewer than 20. The remaining sites were small, having only two and three motifs. By doing the math, it is clear that Mobley’s sample contained a maximum of 212, and probably fewer, rock art figures that he had classified into 43 categories containing at least one element. Even if each class in his typology contained approximately five figures (212 divided by 43), this number is still too small to reveal any correlations among types of figures. Another researcher who has focused on differences in Archaic abstract petroglyphs is Sally Cole.29 Rather than develop a typology of individual motifs—the approach taken by Mobley—Cole identified a West-Central Colorado Abstract Tradition and suggested that these abstract forms can be divided into three general types: 1. discrete elements or motifs, tightly composed (spirals and other designs), 2. indefinite linear elements or motifs, loosely composed and related, and 3. extended linear motifs, tightly composed and related, which may be connected to elements of type 1.30

64

L AWRENCE L. LOENDORF — THUNDER

AND

HERDS

A component of Cole’s typology that is important to a study of High Plains rock art is the identification of a category of discrete elements or motifs—type 1—consisting of figures such as spirals or concentric circles that are independent of connecting images. At one time, Sundstrom made a similar distinction but decided later that discrete forms are so frequently connected by undulating, linear elements that a separate category was unnecessary.31

Early Archaic Rock Art Based on radiocarbon ages and other dating evidence collected at the three localities described in this chapter, type 1 (discrete independent) figures are the oldest elements occurring at Glorieta Mesa. While the inventory there includes many forms that are common elsewhere in Archaic Period rock art, there are also some relatively unique forms, such as starbursts, sunbursts, concentric circles with petals attached to the outer ring, double-ring circles with radiating lines connecting the inner and outer rings, ovals divided by interior parallel lines, circles with long undulating tails, and other circular or ovoid forms (Figure 2.9). At Glorieta Mesa, Archaic individual motif figures—my designation for elements corresponding to Cole’s type 1—share another important characteristic. Although they have been found on the back wall of a nearby rockshelter, these figures occur most often on open-air, horizontal surfaces and, in some instances, incorporate into the design naturally occurring nodules or protuberances in the sandstone. Images resembling bird tracks and figures possibly representing the tracks of ungulates are another significant component of the rock art assemblage (Figure 2.10). From the perspective of research, however, the most important attribute of the Glorieta Mesa petroglyphs is their early age. One might argue that the unique forms at Glorieta Mesa represent a regional variation more common in the southern parts of the High Plains, except for the fact that the individual motif type also occurs on the Uncompahgre Plateau in southwestern Colorado and in the South Dakota Black Hills. Of equal significance is the fact that some of the unique forms at Glorieta Mesa—the doublering circles with radiating lines connecting the inner and outer circles, the starbursts, sunbursts, and concentric circles with petals—are not found at either Clay Creek or the Piñon Canyon boulder sites. The absence of Early Archaic figures at these two sites and in southeastern Colorado generally, and their presence instead in the mountains, high interior basins, and plateaus of Colorado and elsewhere, seem to reflect a pattern of selective land use by foraging peoples during the Early and Middle Archaic time periods. To account for the differences in geographical distribution of rock art prior to the Late Archaic Period, I have developed a model that integrates environmental data with archaeological patterning. It is well established that during the Early Archaic, when the altithermal reduced the availability of water and affected all life forms on the lower Plains,

2 — ROCK ART

OF THE

ARCHAIC PERIOD

65

hunters and gatherers and the animal resources they depended on moved to higher elevations for longer portions of the year. This scenario was first suggested some time ago by James Benedict and other archaeologists, but I am introducing the proposition that if higher elevations were more heavily used in the Early

FIGURE 2.9. Scale drawings of petroglyphs at sites on Glorieta Mesa. Note that most of the figures are independent designs, but a number are connected to other figures. Illustrations by Davíd Joaquín based on field illustrations by A. J. Bock.

66

L AWRENCE L. LOENDORF — THUNDER

AND

HERDS

Archaic than in later time periods, more rock art containing Archaic individual motifs would therefore be located at higher elevations.32 Given the proposed human response to environmental circumstances during the Early Archaic, a related assumption is that at lower elevations, Archaic individual motifs should be located near or directly associated with permanent water sources. Several Plains archaeologists have suggested that Early Archaic

FIGURE 2.10. Scale drawings of petroglyphs at sites on Glorieta Mesa. Note the sequence of bird tracks and the snake-like form, which appear to be representational designs. Illustrations by Davíd Joaquín based on field illustrations by A. J. Bock.

2 — ROCK ART

OF THE

ARCHAIC PERIOD

67

archaeological sites may have become buried near permanent springs or along major river courses where they remain undetected.33 Rock art sites of the time period, however, are much more likely to be more visibly situated, either on the walls of rockshelters or on other exposed rock surfaces. The preceding patterning and assumptions therefore suggest that rock art sites containing Early Archaic individual motif petroglyphs should be more common at elevations exceeding 8000 feet in temperate northern latitudes or, at lower elevations, located along main watercourses or near major permanent springs. Conversely, individual motif elements should be less common in regions where dry, hot conditions prevailed during the altithermal. Although considerably more research is necessary before this model can be tested, it is instructive to look again at the presence or absence of these motifs at the sites discussed in this chapter. I have already noted that figures belonging to the individual motif type are not present in the petroglyph inventory at the Piñon Canyon Maneuver Site, nor do they occur at sites on the eastern Colorado plains such as Clay Creek. While discrete figures are present at Clay Creek, they are rare and almost always connected to undulating lines or other abstract forms.34 Those that do occur, such as the layered lollipop or concentric circles, are bisected by lines that extend beyond the outline of the form. In contrast, such extensions are rare in the Early Archaic individual motifs at sites 6 and 147 on Glorieta Mesa. Petroglyphs belonging to the individual motif type have been found in northeastern New Mexico at Conchas Reservoir and 50 miles to the north at a site near the town of Roy.35 Both locations are within the Canadian River drainage, where Early Archaic occupations occur in rockshelters along the river. The Wagon Mound site, near the New Mexico town of the same name, also contains a series of individual motif petroglyphs. A well-known spring near the town, which was an important stage stop on the Santa Fe Trail, has such a constant volume of flow that it serves as the municipal water source. It is not known if Early Archaic projectile points have been collected in the vicinity of the spring, but a water source of this size and reliability was likely to have been a magnet location during parts of the altithermal. The chronology of the Wagon Mound rock art site is complicated by the fact that it contains dozens of Middle and Late Archaic petroglyphs, as well as figures from more recent time periods. The spatial distribution of the petroglyphs, however, provides a clue to their age. Later figures are primarily clustered on the basalt cliff face that traverses the site, while the earlier individual motif figures occur on basalt boulders near the spring and the stream that flows from it. Although eroded and difficult to see, the individual motif elements consist of still-recognizable concentric circles, starbursts, and other figures found at Glorieta Mesa.36 My model of the distribution of Early Archaic sites containing individual motif petroglyphs includes the expectation, which is met by Glorieta Mesa, that

68

L AWRENCE L. LOENDORF — THUNDER

AND

HERDS

they would be located at elevations higher than 8000 feet. The discovery of Early Archaic projectile points on the mesa supports the model, even though, to my knowledge, none have been recorded from excavated contexts. A broken dart point found on the surface of one petroglyph panel appears to belong in the Early to Middle Archaic time frame. It is also likely that any sites in central Colorado that contain individual motif petroglyphs will be located at higher elevations and associated with evidence of Early Archaic occupations. At least one example of an Early Archaic presence in the mountains consists of archaeological remains of a hunting and gathering adaptation referred to as the Mount Albion Complex.37 High-altitude sites of this type in the Rocky Mountains have been dated to between 5800 and 5300 years ago. Known more for its association with game drives, the complex has yet to be linked to any rock art sites. One possible exception is a single petroglyph on James Peak, at an elevation exceeding 11,000 feet, which is similar to Early Archaic individual motif figures. The James Peak figure is a circle with interior, parallel, ladder-like lines and radiating, exterior, wavy lines that make it look very much like a sunburst. The image, originally found by hikers, is engraved on a granite boulder that is nearly covered by vegetation. Unfortunately, recent attempts to relocate the boulder have not been successful. Based on archaeological evidence, individual motif petroglyphs in the Black Hills occur in an area that may have been more temperate during the Early Archaic. Archaeological remains from the Hawken site, a 6000-year-old bison kill site, provide evidence that during parts of the altithermal the region was well watered and would have been an excellent habitat for bison and the humans who hunted them.38

The Middle Archaic Rock art from the Middle Archaic time period—between 5000 and 3000 B.P.— occurs at Clay Creek and other sites on the High Plains, as well as at several Piñon Canyon Maneuver Site localities (Figure 2.11). During this 2000-year period, petroglyph figures resemble elements that Cole has designated as type 3: “extended linear motifs, tightly composed and related, which may be connected to elements of type 1.”39 In other words, Early Archaic individual motifs continue to be present in the Middle Archaic, but instead of being unattached, they are more often connected to one another by undulating lines or may actually overlap. As a rule, the pecked lines of Middle Archaic individual figures are less precise and less deeply pecked than in the Early Archaic. There is more variability in undulating or wandering lines, which can vary in width from 1 to 4 centimeters, and interior lines often extend beyond the outline of individual motifs. For example, in Early Archaic figures an interior line that bisects a circle usually terminates at the circle’s outline, while in the Middle Archaic such a line is more likely to extend outside the circle on one or both sides.

2 — ROCK ART

OF THE

ARCHAIC PERIOD

69

More research is needed to determine which Early Archaic individual figure types are no longer present in the petroglyph inventory of the Middle Archaic and which new types are introduced in that time period. One new Middle Archaic form appears to be concentric circles that have double parallel lines around the perimeter. The best examples are found at Clay Creek, but these figures also occur at the Bent Canyon stage stop site. In the historic era, the freshwater spring at Bent Canyon provided water for the horses and drivers of the Barlow and Sanderson Southern Overland Mail and Express Company which operated from 1871 until 1876, when newly introduced rail service made horse-drawn transport obsolete.40 The spring is surrounded by archaeological sites dating from its use by stage stop patrons and— several thousand years earlier—Archaic-era foragers. Several rock art sites in the vicinity include a large sandstone boulder on which are engraved rayed circles, bisected grids, lines with bisecting horizontal crossing arms, and other forms similar to those occurring at Clay Creek. Nearby are several linear figures consisting of wide, pecked vertical lines that end in a circle, like a do-not-enter sign. The opposite end of one of these designs consists of an inverted V, which might tempt a contemporary viewer more inclined to see a realistic rather than an abstract image to interpret the figure as an anthropomorph (Figure 2.12).

FIGURE 2.11. Abstract petroglyphs from the Hicklin Springs site in southeastern Colorado. Note the quadruped-like figures in the lower left and the anthropomorph-like figure in the lower right. The panel measures about 4 meters across. Drawing by Chris Alford based on a field drawing by Peter Faris.

70

L AWRENCE L. LOENDORF — THUNDER

AND

HERDS

FIGURE 2.12. Petroglyphs from 5LA9781, a site near the Bent Canyon Stage Stop, Piñon Canyon. This site is Archaic in age, but it contains figures identified as anthropomorphs by some researchers and as simple abstract forms by others. Scale drawings by Terry Moody.

The Late Archaic During the Late Archaic time period—between 3000 B.P. and 1850 B.P.— abstract figures were inscribed on surfaces in a wide variety of site types on the High Plains. New designs, and new interpretations of older designs, appear during this period. It is, however, sometimes difficult to distinguish between Middle Archaic and later figures because many petroglyphs are either in close proximity to or overlap one another on sandstone walls. The Piñon Canyon boulder sites are among the few localities where the study of Archaic Period rock art chronology is simplified by the placement of petroglyph figures on individual boulders, which separates one petroglyph-making event from another. The transition from Middle Archaic oval-shaped grids to more rectangular Late Archaic grids is a particularly noticeable example of the change from curvilinear to rectilinear figures. Corners or sides of abstract figures that were once rounded also become squarer. A more definitive characteristic of Late Archaic sites, however, is the presence for the first time of quadruped figures (Figure 2.13). Found singly or in pairs, these quadrupeds are invariably simple figures with straight legs that end without hooves. They may have head appendages but are never shown with elaborate branching antlers. Quadrupeds are often connected to abstract forms, and occasionally they will form a line or small group, as though they are part of a herd. Researchers are fairly certain that bird figures are also included in the Late Archaic inventory of representational forms, but because these petroglyphs occur

2 — ROCK ART

OF THE

ARCHAIC PERIOD

71

FIGURE 2.13. Quadruped figure on a basalt boulder at the Hogback Archaic site, Piñon Canyon Maneuver Site. Stiff-legged quadrupeds are assigned to the Late Archaic Period. Drawing by Chris Alford based on the field drawing by Seija Karki.

in much the same form for the next 1000, at present it is not possible to assign any particular bird figure to the Late Archaic Period. Images of birds are known to occur, however, among exclusively Late Archaic figures at two localities at the Piñon Canyon Maneuver Site and at one site in the nearby Chacuaco area of Las Animas County, Colorado. In each instance, the fully frontal bird figures have the straight body, fan-shaped tail, and outspread wings of the classic thunderbird of later Plains Indian art. Simpler but related images of bird tracks have been recorded in the Early Archaic Period and continue throughout the Middle and Late Archaic. They are sometimes found in vertically arranged pairs, but such an alignment, which suggests walking, is not known to include more than two images. No distinctly anthropomorphic figures have been assigned to the Archaic Period on the High Plains, although suggestively human-like forms, such as the linear figure at the Bent Canyon site which has a head-like circular form at one end and an inverted V that might represent feet at the opposite end, continue throughout the Late Archaic. Figures of this type are sufficiently abstract to cast doubt on any assertion that they are meant to represent humans (Figure 2.14).

A LOOK BACKWARD At this juncture I had expected to continue the temporal trajectory established in this chapter and proceed from a consideration of the more distant rock art record on the High Plains to that of the more recent past. Looking forward in

72

L AWRENCE L. LOENDORF — THUNDER

AND

HERDS

FIGURE 2.14. Vertical figures at the Hogback Archaic site, Piñon Canyon Maneuver Site. Some researchers have identified the upright figure as an anthropomorph. Note the figure's similarity to the designs at 5LA9781 illustrated in Figure 2.12. Drawing by Chris Alford based on the field drawing by Seija Karki.

time, I had intended to discuss how the directional trends that are apparent in Late Archaic sites had continued in the Late Prehistoric Period, as more—and more elaborate—quadrupeds are depicted and the earliest unambiguous anthropomorphs appear. The results of recent research at the Piñon Canyon Maneuver Site, however, have prompted me to temporarily reverse chronological direction. It has long been thought that the rock art record of the Archaic Period, which ends at approximately 1850 B.P., represents the initial and oldest body of rock art images on the High Plains. The view that Archaic Period rock art is seminal may be incorrect, however, and in need of revision. A look backward, based on recent evidence, may in fact reveal that Paleo-Indian rock art precedes the Archaic inventory of the High Plains.

ROCK ART

OF THE

PALEO -I NDIAN PERIOD

The Paleo-Indian Period is the name given to the thousands of years during which peoples from Asia first entered and spread throughout the Americas.41 Just when the migration began, and over what period of time it occurred, is the subject of debate among archaeologists, but there is consensus that the end of the period coincides with the termination of the Pleistocene era and the extinction of Ice Age megafauna in North America. The Paleo-Indian Period is subdivided into three successive temporal units often referred to as the Clovis, Folsom, and

2 — ROCK ART

OF THE

ARCHAIC PERIOD

73

Plano cultures, named for the sites on the High Plains where distinctive stone tools associated with the remains of extinct animals were first found. The initial discovery of Folsom projectile points embedded with the bones of Bison antiquus occurred at a site east of Raton, New Mexico. Projectile points associated with Mammuthus columbis were found at the Blackwater Draw site near Clovis, New Mexico, and evidence of a Plano Period communal kill of Bison antiquus was recovered from the Olsen-Chubbuck site in southeastern Colorado. Because these and many other sites of the period contained the bones of large herbivores that had been killed and butchered, Paleo-Indian peoples have often been referred to as Big Game Hunters. It is not surprising, then, that since large mammals appeared to represent a significant component of the diet of First Americans, efforts to find Paleo-Indian rock art have been directed toward the discovery of pictographs and petroglyphs containing images of mammoths, bison, and other large animals. The many Upper Paleolithic paintings and engravings of mammoths and bison in European caves have been used to support the assumption that New World hunters would also have painted, pecked, or incised rock art images of these species. Similar expectations about Paleo-Indian rock art continue to be reflected in American rock art literature. In a recent book titled The Settlement of the American Continents: A Multidisciplinary Approach to Human Biogeography, Larry Agenbroad and India Hesse argue, contrary to the current consensus, that the Colorado Plateau was populated by big game animals during the Paleo-Indian Period.42 The authors underscore their argument by identifying the localities where the bones of extinct mammoths, bison, and other Pleistocene fauna have been recovered and comparing these to the locations where Paleo-Indian projectile points have been found. They also point to pictographs and petroglyphs of bison and mammoth-like figures, although they note that some of the images could be representations of circus elephants. The possibility that the images might be fraudulent does not, however, shake their conviction that mammoths and bison should be present in the Paleo-Indian rock art repertoire, because— the argument goes—these are the animals that Paleo-Indian peoples hunted. I have to confess that I was operating with a similar set of expectations when I embarked on a search for Paleo-Indian rock art on the central High Plains. I began by looking for images of mammoths or bison in an area that contained many large outcrops of basalt because, I reasoned, ancient petroglyphs in exposed settings would be best preserved on these very hard sediments. Then I narrowed my search and began to look for Archaic-age sites whose distinctive abstract petroglyphs I knew I could identify. I was looking specifically for any Archaic petroglyphs that might have been superimposed over petroglyphs of mammoths or bison, and because I was looking in places within a few kilometers of the original Folsom site, I was hopeful that I would be successful.

74

L AWRENCE L. LOENDORF — THUNDER

AND

HERDS

I had nothing to show for my efforts when, one day in 2006, La Junta, Colorado researcher Mike Waugh was looking for rock art sites with me and noticed something unusual. Mike’s sharp eyes had spotted several old-looking Archaic abstract pecked petroglyphs at a site on the Piñon Canyon Hogback that we have since named “Ancient Hogback.” The petroglyphs were so heavily varnished they were nearly indistinguishable from the background color of the basalt, unless looked at from an oblique angle and in just the right light (Figure 2.15). At first I was disappointed to discover that there were no images of mammoths, bison, or other large prey animals under the very old pecked figures. Instead, we found a series of incised abstract designs, one of which was a zigzag that reminded me of the 10,000-year-old bison skull discovered at the Cooper site in western Oklahoma. At this Folsom-age site, on three successive occasions, herds of Bison antiquus had been driven into a dead-end arroyo and killed by hunters waiting in the gully.43 At some point after the first drive, a red zigzag design resembling lightning had been painted on a sun-bleached skull that was then placed by Folsom hunters in the arroyo, possibly as a talisman to attract a second group of bison. Another figure at the Ancient Hogback site consists of an incised oval with two pointed ends that is oriented vertically and bisected by a vertical line. The incised vertical line extends through the figure’s pointed ends, forming fringe-like tassels

FIGURE 2.15. Laurie Lee, a member of the Colorado Rock Art Association, shown recording one of the petroglyph boulders at the Ancient Hogback site, Piñon Canyon Maneuver Site. Photograph by Lawrence Loendorf.

2 — ROCK ART

OF THE

ARCHAIC PERIOD

75

on top and a zigzag below that resembles lightning. Three horizontal lines cross the vertical element midway between the two ends of the oval (Figure 2.16). Some viewers have commented that the figure looks like a rattle, but this assessment can neither be verified nor rejected. No abstract pecked figures overlie this panel, but incised images in a neighboring panel clearly underlie Archaic-age petroglyphs. These include ladder-like forms and four figures that look like atlatls or arrows with feather fletching, as well as the zigzags that Mike and I first discovered at the site. Double parallel lines were used to outline many of these figures (Figure 2.17).

FIGURE 2.16. A tracing of the incised figure on Boulder 1 at the Ancient Hogback site. Some researchers have identified this figure as a rattle. Field tracing by Kendra Rodgers.

FIGURE 2.17. A tracing of the incised figures underlying abstract pecked images. Note the small arrow-like designs. Field tracing by Kendra Rodgers.

2 — ROCK ART

OF THE

ARCHAIC PERIOD

77

More Questions Than Answers My examination of the figures at the Ancient Hogback site has led me to review other figures that are believed to represent Paleo-Indian rock art in North America. A petroglyph in the Mojave Desert that resembles a bighorn sheep has recently had its age revised to 12,000 years old, a time when bighorns were common to the region.44 Another possible Paleo-Indian petroglyph appears to depict a Mojave Desert llama, a species that has been extinct in the region for 7000 years. David Whitley, working with Ronald Dorn and Tanziu Liu, has dated this figure by cation-ratio at 13,400 ± 2000 B.P. Varnish microlaminations for the figure indicate it was made between 12,500 and 16,500 B.P. Still a third dating technique was used to date a mineral skin or calcium oxalate coating on the petroglyph with an age of 11,860 B.P.45 Even though these dating methods are experimental, the confirmation of all three ages for a single petroglyph is significant. They suggest that this petroglyph, whether it represents an extinct llama or not, is very old. Some figures of animals in Wyoming’s Black Hills may be 11,500 years old, and quadruped images at Legend Rock, Wyoming, have cation-ratio ages in the 10,000-year range.46 Although the quadruped figures at Legend Rock and at sites in the Black Hills are probably the oldest rock art images in their areas, until better dating techniques become available there is still a question of whether they were produced by Paleo-Indian or Archaic peoples. The quadruped petroglyphs in the Mojave area, which are also very old, are roughly contemporary with nearby abstract forms, so that it remains unclear which figure types are older. At this juncture I also have to admit that none of the reputedly oldest figures are either mammoths or extinct bison, but instead are either a possible llama, undifferentiated quadruped species, or examples of bighorn sheep, elk, or deer. And none, I might add, are made by incising, the technique used to make the old abstract images at the Ancient Hogback site.

Temporal Patterning of Incised Stones Excavations begun in the 1990s in Clovis-age deposits at the Gault site in Williamson County, Texas, have resulted in the discovery of more than two dozen stones and pebbles with incised abstract motifs.47 The checkerboard designs and other abstract patterns, made with both single- and double-width lines, clearly establish that Paleo-Indian peoples did indeed make abstract figures on portable stone objects. Some researchers claim to be able to see representational elements on the Gault stones, but not even the most open-minded researcher has suggested that any of the figures represent mammoths or bison. Archaeological opinion has, in fact, coalesced around the view that the patterns are clearly more abstract than representational.

78

L AWRENCE L. LOENDORF — THUNDER

AND

HERDS

The inventory of stone artifacts incised with abstract figures also includes those from the 9400-year-old Paleo-Indian site at Barton Gulch, Montana, whose incised lines of differing depths formed cross-hatched designs.48 At the Long Lake site in Oregon, a petroglyph with rows of dots and dashes was covered by sediments containing Mazama ash, a well-dated soil formation that is used extensively in the western United States as a relative dating tool. If the soil layers covering the Long Lake panel have not been mixed in some manner, the abstract petroglyphs they cover are more than 6700 years old.49 Based on these examples, there appears to be a linkage between incised abstract designs and Paleo-Indian peoples, but other evidence points to the fact that this association is not exclusive. For example, during a study of mobiliary art at the Boca de Potrerillos site in Nuevo Leon, Mexico, archaeologist Solveig Turpin and her colleagues found a surface scatter of incised plaquettes around hearths dated to between 4820 and 5460 years ago.50 Surface artifacts associated with the incised tablets included Clear Fork tools that are minimally 3000 years old. Turpin has noted that while the tablets at Boca de Potrerillos are probably Early Archaic in age, incised stones from all time periods have been found in sites across southern Texas. Incised stones have also occurred in a wide variety of contexts and in all time periods in the Great Basin and adjacent regions. They have been found in cultural layers dated to older than 5000 years at Hogup Cave, near Salt Lake City, Utah, and they have also been recovered from numerous excavated contexts of more recent age.51 One of the best controlled studies of incised rock art in the Great Basin was conducted by Eric Ritter at the Pistone and Massacre Bench sites in Nevada. Ritter believes that although much of this rock art was created in the past 1000 to 1500 years, some may date to the Middle Archaic.52 He concludes that the placement, over millennia, of incised abstract designs on rock surfaces reflects a tradition in which individuals experienced altered states of consciousness during which they observed and later recorded entoptic phenomena. Whether this explanation accounts for the presence of incised abstract figures, both mobiliary and parietal, either in Nevada or elsewhere, has not, however, been sufficiently evaluated.

Present But Not Definitive Initially it should be clear that there may be quadrupeds of Paleo-Indian age in the Mojave as well as Wyoming. All that I can speak toward here are the images I have found on the central High Plains. It should also be clear that the presence of incised abstract designs underlying the pecked Archaic-age petroglyphs at the Ancient Hogback site are not necessarily Paleo-Indian in age. The suite of figures may just as easily have been produced by other Archaic peoples and, in fact, may have been created only a few minutes before pecking began on the overlying designs. Of course, one could ask rhetorically why someone making an incised

2 — ROCK ART

OF THE

ARCHAIC PERIOD

79

design would abruptly change gears and begin making a pecked one on top of it. That far-fetched scenario can be definitively rejected simply by pointing out that heavier varnish is present on the underlying incised figures than on the pecked designs that overlie them, proving that a significant amount of time elapsed between the creation of the two sets of images. Another important consideration is that incised abstract figures usually cooccur and are contemporary with representational rock art. Ritter has reported that a few images of incised animals were intermixed with pecked representational figures in Nevada. Similarly, on the central High Plains, grid-like patterns and circular designs are associated with Developmental Period quadrupeds, although it is not until the Historic Period that the incised figures of horses and bison occur. The absence of representational figures at the Ancient Hogback site is significant, particularly since the abstract figures there are typical of Early or Middle Archaic pecked designs that I discussed earlier in this chapter. I still believe there is a very real possibility that the Ancient Hogback incised figures are Paleo-Indian in age and that there may be some very old incised rock art that has so far been overlooked at sites on the central High Plains. Even so, I haven't yet given up hope that Paleo-Indian peoples made rock art images of mammoths, mastodons, and bison as well as incised abstract figures and that someday we will find them. I would be the first to admit, however, that definitive evidence to the contrary may become available, at which point I will have to release my cherished expectations. Until that day comes, though, it is not irrational for me to keep looking.

NOTES 1 Colorado Springs Free Press, Sunday, June 20, 1965, “LBJ Makes Declaration as Arkansas River

Rages,” p. 1.

2 These and other rainfall data can be found at the Website of the Western Regional Climate

Center, http://www.wrcc.dri.edu/cgi-bin/cliGCStP.pl?colama.

3 The Lamar-Tri-State Daily News, 25th Anniversary Supplement, Tuesday, June 17, 1990,

“Holly Remembers the ’65 Flood,” p. 7. Buckles 1982:3. Buckles 1980; James 1983. Buckles 1980; 1989:128. Buckles 1982:5. Trimble 1980. This and other information about Colorado’s glacial history is found at the Glaciers of the American West website, http://glaciers.research.pdx.edu/states/Colorado.php#Glaciated_Regions. 10 Loendorf 1989:148. The cation-ratio ages for the boulder surfaces are 14,900 ± 1150 B.P. and 14,000 ± 550 B.P. 11 Andrefsky et al. 1990:1014; Loendorf 1989:166–67. 12 Loendorf 1989:167. 13 Anell 1969:48–49, 74. 4 5 6 7 8 9

80

L AWRENCE L. LOENDORF — THUNDER

AND

HERDS

14 A detailed discussion of the events surrounding the battle of Glorieta Pass can be found in

Edington and Taylor 1998:4.

15 The official site number is AR-03-10-05-006. 16 The official site number is AR-03-10-05-147. 17 Abel 1994. 18 Bock and Bock 1992, 1993, 1994. 19 Bock and Bock 1994. 20 Phillips 1994. 21 Dorn 1993, 1994; Abel 1994. 22 A photograph of the projectile point appears in Bock and Bock 1994:124. 23 Phillips 1994. 24 Dorn 1993, 1994; Abel 1994. 25 Sundstrom 2004:68–77; P. Schaafsma 1980:43–47. 26 Heizer and Baumhoff 1962; Turpin et al. 1996. 27 Cole 1987:39; 1990:84. 28 Mobley 1981:145–76; Steed 1984. 29 Cole 1987:35; 1990:49–50. 30 Cole 1987:37. 31 Sundstrom 2004:73. 32 Benedict 1979; Black 1991. 33 Meltzer 1999. 34 Buckles 1989:129. 35 P. Schaafsma 1992:140. The Roy site is not reported in the literature, but during a visit to the

site in 2007 I found that it contained the same independent motifs that occur at Glorieta Mesa. 36 The Wagon Mound site was reportedly recorded by the Taos Archaeological Society, but to my knowledge a report has not been issued. 37 Benedict and Olson 1978 38 Frison 1991:187–91. 39 Cole 1987:37. 40 Church and Cowen 2003. 41 Zier 1999b:73–99 presents an up-to-date overview of the Paleo-Indian Stage in southeastern Colorado. 42 Barton et al. 2004. 43 Bement 1999:37–40. 44 Whitley and Dorn 1993. 45 Whitley 1999, 2008. 46 Tratebas 1993; Sundstrom 2004; Francis and Loendorf 2002. 47 Collins et al. 1992. 48 The Barton Gulch engraved stones are not reported in the literature. 49 Ricks 1995. 50 Turpin et al. 1996. 51 Aikens 1970; James 1983. 52 Ritter 1994.

3 PURGATOIRE PECKED-I PETROGLYPHS

he alliterative taxon of central High Plains rock art known as Purgatoire Pecked-I Petroglyphs was originally identified and named in the 1980s by researcher Sally Cole.1 More recently, extensive research has refined our knowledge of the temporal boundaries and spatial distribution of this recognizable group of figures. Quadrupeds are the most frequently represented figures in the Purgatoire Pecked-I Petroglyph style. They include five, and perhaps six, species of ungulates. Whitetail deer, mule deer, pronghorns, and bighorn sheep are common figures, while bison, and quadrupeds that may represent elk, occur in low enough frequencies to qualify as occasionally present. Other identifiable animals include bears, figures resembling dogs or coyotes, and snakes. Purgatoire Pecked-I anthropomorphs are distributed in and among the quadruped figures. These human forms vary from simple stick figures to the more frequently depicted static, front-facing, full-view figures that have elongated, ovoid bodies and outstretched arms ending in digitate hands. The style includes a fairly substantial inventory of abstract forms such as circles and modified or augmented circular forms. At some sites, long pecked lines are attached to animal figures or may simply undulate and twist across a part of a panel. In recent years, I and my team have identified grid-like figures that we now refer to as nets. These are somewhat variable in form and may, in fact, represent a range of hunting paraphernalia and features such as snares, corrals, and other kinds of traps.

T

81

82

L AWRENCE L. LOENDORF — THUNDER

AND

HERDS

Purgatoire Pecked-I Petroglyphs are associated with the Developmental Period (Table 1.1). The designation “developmental” has been assigned to this time period to indicate that not only did characteristics of the archaeological record first appearing in the Late Archaic become much more prevalent as time passed, but also that important innovations in material culture occurred.2 A look at residential continuity and change in the Developmental Period reveals that people continued to live in rockshelters and caves, but that increasingly they occupied open-air sites containing semi-permanent dwellings. Their new oval structures consisted of brush or grass superstructures that rested on foundations made of upright sandstone slabs or masonry walls. These structures contained several kinds of fire-pit features, including ones used for roasting, and pit features dug into the floor that may have been used for storage. A significant technological innovation is signaled by the presence on Developmental sites of small, corner-notched Scallorn arrow points which reflect the adoption of the bow and arrow in some hunting situations. Their presence side by side with large corner-notched projectile points used to tip spears during the Late Archaic, however, demonstrates that hunters used a mix of new and older hunting techniques. The chipped stone inventory also included scraping and cutting tools that are very similar to earlier forms. Implements used primarily in domestic settings also reflect a mixture of continuity and innovation. Cord-marked, conical-shaped vessels first appear in small numbers in the artifact inventory at this time. Numerous one-hand manos, used extensively in the Archaic Period for grinding wild seeds, continue to be found in the Developmental Period in association with basin, slab, and bedrock metates. The latter differ from the typical trough metates commonly used for processing maize in other parts of the West. Although maize from small garden plots has been found in some archaeological deposits in the region, many sites contain no evidence of horticulture. The picture created from surface collection and excavation suggests that subsistence continued to be based primarily on gathered plant resources and hunted animals, supplemented by some food production and involving a combination of lithic technologies. Although there are numerous sites in the region that contain Purgatoire Pecked-I petroglyphs, those I have chosen to discuss in this chapter reflect particularly well the variety that occurs in the characteristics of this style.

THE BULL PASTURE SITE Many of the accounts summarizing the prevalence of rock art in a particular area— which can be as large as the state of Texas or as small as the county of West Sussex in England—contain a statement in the first paragraph noting that “most rock art sites

3 — PURGATOIRE PECKED-I PETROGLYPHS

83

are located on privately owned land.” Sometimes these statements are followed by a caveat warning that gaining access to such rock art sites can be extremely difficult. What does a researcher do when he or she wants to study all of the rock art in a region and not just the sites that are located on public land and managed by governmental agencies? I have been fortunate throughout my research on the High Plains to have had access to an extraordinarily complete set of photographs and records of privately owned regional rock art sites assembled and maintained by Nancy Robertson, a resident of Raton, New Mexico. Luckily for me, Nancy has shared her archive with qualified researchers, so I have been able to bypass the question of site ownership and grapple immediately with problem number two: getting permission from landowners, who are mostly ranchers, to visit their sites. An understanding of etiquette as it applies to ranching life is essential to success and usually begins with the tactful timing of a telephone request for site access. In the spring, ranchers are busy with calving, and from July through September they are almost certainly irrigating or haying, which means they go to work at sun-up and come home after dark. I don’t feel comfortable calling a tired rancher after 8:00 in the evening, and I have learned that they are more receptive to intrusions between 6:00 and 7:00 in the morning. The voice that answers the phone often belongs to a gruff-sounding male who wants to know how I learned about the rock art site on his property. When I explain that my source is Nancy Robertson, a common response is, “She ain’t supposed to be telling anyone about those Indian writings!” I always reply that Nancy probably told me because I am a qualified researcher, to which the voice usually responds, “We don’t want anyone knowing about that place.” I explain that I will not give anyone any information about the site’s location, nor will I will take anyone to the site. Most importantly, I tell the rancher that I will keep him informed about anything that I see or do at the site. Following a short sermon instructing me to be “damn sure you don’t go hauling a bunch of people out there,” I am usually given permission for a site visit. Once I have returned to my office, I always write to the rancher, thanking him for giving me access and explaining my thoughts about the rock art. And I always make sure that I express my appreciation for his vigilance in protecting the rock art. Ranchers are a close-knit group. They talk to each other about outsiders, and in a short time any person requesting access to rock art sites on private land within a fairly wide region has acquired either a good or a bad reputation. Because of their careful monitoring and information-sharing about the rock art sites on their land, ranchers are outstanding custodians of these irreplaceable resources, and their sites are among the best protected in North America. A case in point is the pristine condition of archaeological materials ranging from intact pit houses to storage facilities to rock art panels at the Wilcox Ranch on Range Creek in

84

L AWRENCE L. LOENDORF — THUNDER

AND

HERDS

Utah, which made a folk hero of ranch owner Waldo Wilcox. When Wilcox transferred ownership of his 4200-acre spread to the state of Utah, one archaeologist labeled the gift a “national treasure,” and another commented that “it’s like finding a Van Gogh in your grandmother’s attic.”3 After one of my requests for permission to visit a rock art site had been granted, the rancher added that I must remember to close the gates and to “be careful of them bulls down there.” A few days later, I made my way to the site, which is close to Tinaja Peak in north-central New Mexico. When I arrived, I was immediately impressed by the many quadruped petroglyphs, in several different sizes, which were distributed over a series of basalt boulders. Once part of a ridge that had formed during a region-wide ancient lava flow, the boulders had separated from their parent bedrock during episodes of freeze-thawing and had rolled down the slope. Several were large, with upright exposed flat sides that were 2 meters high and 2 to 3 meters wide. On one boulder, I counted more than 40 quadruped figures, and there were 20 more on another (Figure 3.1). By the time I had looked at all of the petroglyph-covered boulders, I had counted at least 250 quadrupeds, both small and large, representing the Purgatoire Pecked-I style. The figures could be grouped into three size classes. The smaller animals tended to measure about 9 centimeters from head to tail, while larger ones were

FIGURE 3.1. A few of the quadruped figures on a boulder at the Bull Pasture site. Note the variability in the size of the figures. Photograph by Lawrence Loendorf.

3 — PURGATOIRE PECKED-I PETROGLYPHS

85

usually slightly more than twice that size. A few outsized quadrupeds measured 35 to 40 centimeters from head to tail and were invariably depicted with large sets of antlers. In place of antlers, many of the mid-size figures—probably females—had head appendages that look like ears. The smallest figures—in the 9-centimeter range—probably represented juvenile animals. Viewed as an assemblage, the figures appeared to represent one or more deer herds. It was at the Bull Pasture site that I first saw the vertically oriented, ladderlike forms that occurred between groups of animals. Experience has since taught me that these figures are common at other sites in the region from the same time period, and I am now fairly certain that they represent hunting nets. Interspersed among the animal figures at the Bull Pasture site are abstract forms such as circles, circular figures, and a few wavy lines, but these images are significantly outnumbered by the much more abundant quadrupeds. I was marveling to myself about the number and variety of the quadrupeds at the site when I noticed anthropomorphic figures on a boulder on the opposite side of the fence from the main site. Focusing on these intriguing images to the exclusion of my surroundings, I scrambled over the fence and headed for a closer look. I had almost reached the boulder before I realized that, in my excitement, I had inadvertently invaded the bull pasture, where six or seven large Angus bulls were standing in a grove of trees, a dozen or so meters from the boulder. The western admonition “never trust an Angus bull” came to mind as I stopped in my tracks and debated whether the prospect of finding some interesting rock art figures was worth the risk of serious injury. Fortunately, the bulls seemed more interested in remaining in the cool shade than in expressing their dominance over a human interloper but, even so, I did not spend as much time examining the rock art figures as I might have under different circumstances. My somewhat cursory inspection revealed that three anthropomorphs, five or six quadrupeds, and two circles had been inscribed on the boulder. The anthropomorphs were full-facing figures with linear bodies, outward curving legs ending in L-shaped feet, and straight arms without hands. Long penis-like appendages between their legs suggested they were male. One of the anthropomorphs was more robust than the other two, which were stick-like. The sturdier anthropomorph extended an object composed of concentric circles toward a quadruped, to which the quadruped appeared to be reacting by standing on its hind legs. A similar circular object was placed between the two stick-like human figures (Figure 3.2). While these forms have been interpreted by other researchers as possible shields, there are several reasons why this identification may be incorrect.4 My first concern has to do with the objects’ uniqueness. I have now examined more than 80 sites containing Purgatoire Pecked-I petroglyphs, and in none of them is a similar object held by a human figure. If the image was meant to depict a shield, one would expect to find other examples of shields in the rock

86

L AWRENCE L. LOENDORF — THUNDER

AND

HERDS

FIGURE 3.2. An anthropomorph at the Bull Pasture site holding a hoop toward a quadruped. Photograph by Lawrence Loendorf.

art at other locations. My second concern relates to the body language of the anthropomorphic figure holding the round object. There is no suggestion of aggression toward the quadruped, nor does the human figure assume a defensive posture, which would be typical of a shield-holding warrior. The most convincing argument against interpreting the concentric circle design as a shield is based on the presence of similar figures at other sites, intermixed with quadrupeds but not held by humans. In such settings, these elements appear to derive their meaning from their proximity to prey animals. This association may suggest that the circular object in the hand of the anthropomorph at the Bull Pasture site is some sort of shaman’s implement, such as a magical hoop used for charming deer, rather than a shield whose purpose would have been self-defense.

THE ZOOKEEPER SITE Hap is a noun seldom used in contemporary speech. According to the Oxford English Dictionary, however, it occurred frequently in the discourse of medieval Britons and meant “chance or fortune (good or bad) that falls to any one; luck, lot.”5 Hap is also the operative syllable in the verb to happen, which scientists, including archaeologists, often use to indicate that their first encounter with the phenomena they study was unplanned. For instance, paleoanthropologist Donald Johanson has described in the following terms his discovery of the famous and important fossil named Lucy: “I was heading back to my Land Rover to drive back to camp. And I happened to look over my right shoulder. And as I did so, I saw a

3 — PURGATOIRE PECKED-I PETROGLYPHS

87

fragment of a bone which I recognized as coming from the elbow region in a skeleton, and that it was too small to be anything but one of these Hominids.”6 The list of well-known archaeological sites that just happened to be discovered is long and includes the rock art site of Lascaux, many Maya ruins, and the life-size Chinese terracotta figures unearthed in 1974 by local farmers drilling a well near the city of Xian. Hap, or chance, was in the air in the early autumn of 1988 when Dr. Orrin Rongstad of the University of Wisconsin brought a group of students to southeastern Colorado to observe the projects being conducted at the Piñon Canyon Maneuver Site (PCMS). One afternoon, Rongstad took a number of students to visit Picket Wire Canyon, a protected site maintained by the U.S. Forest Service. Located in the Purgatoire River Valley on the northeastern boundary of PCMS, it contains the largest dinosaur-track site in North America. Conditions in the canyon were very different 150 million years ago; in fact, neither the Rocky Mountains nor the canyon yet existed. The western landscape then consisted of nearly flat terrain interspersed with lakes, streams, and the abundant vegetation that appealed to plant-eating apatosaurs and provided habitat for the prey of meat-eating allosaurs. The Purgatoire River Valley itself had not yet emerged and was instead subsumed in a vast lake, along whose shores numerous dinosaur species traipsed. They left their footprints in muddy sediments that, over millennia, were transformed into sandstone bedrock containing evidence of the passage of the now-extinct creatures.7 Parking their vehicle at a fence on the western rim of the canyon, some students hiked down a steep slope to the river bottom, passing the Sorenson site on their way. As others followed the canyon’s rim in their descent, their path took them by a cluster of rock art images that has come to be known as the Zookeeper site. One student stopped to photograph a large petroglyph panel which, like Donald Johanson in a different time and place, he happened to notice before continuing north toward the long lines of tracks—over 1300 in all—embedded in the exposed rock of the canyon floor. Unaware that he had just made an important discovery, the student assumed that his photograph had recorded an already known site, and he didn’t mention the rock art panel to his companions. It wasn’t until several months later, when the team reassembled in Madison to share photographs and to discuss the experiences of the preceding summer’s fieldwork, that the significance of the chance encounter became clear. Orrin Rongstad wasn’t present for the photographic show-and-tell, but David Anderson, the wildlife biologist at Piñon Canyon Maneuver Site at the time, was, and he realized that he was looking at a significant and previously unknown site.8 In due course, Steve Chomko, at that time the archaeologist responsible for fieldwork at Piñon Canyon, was contacted. Chomko initially recorded the site and was later joined by other archaeologists who participated

88

L AWRENCE L. LOENDORF — THUNDER

AND

HERDS

in the site’s more intensive documentation.9 As a description of its figures and composition will demonstrate, the location is aptly named the Zookeeper site and well deserves its reputation as one of the most impressive rock art sites on the High Plains.

The Petroglyphs at the Zookeeper Site The main panel at the Zookeeper site is located in a shallow alcove approximately 1 meter above the surrounding ground surface. The alcove, which is nearly 6 meters wide and 3 meters high, is recessed into a sandstone wall at a depth of about 1 meter. Large juniper trees grow on the bedrock bench in front of the alcove but do not obscure the petroglyph panel from view. Thirty-six solidly pecked quadruped figures surround the single anthropomorphic figure at the approximate center of the panel and create a dominating presence. Although the official site name is 5LA5993, the positioning of so many impressive animal figures around one central human-like figure—which suggests that they were meant to be part of the same scene—underscores the appropriateness of its vernacular name, the Zookeeper site (Figure 3.3). From head to toe the zookeeper figure measures about 15 centimeters, but its horned headgear adds another 5 to 7 centimeters to its height. Shown in fullfrontal view, the figure has an elongated, ovoid body to which an inverted set of U-shaped legs is attached. Its straight arms project away from the body, and each ends in four disproportionately large, splayed fingers. Attached to the anthropomorphic figure’s right hand is a set of grid-like crossed lines that are larger than the figure itself. The object’s frayed ends suggest it might be a net or snare—or some sort of object of power—but it is clearly not an atlatl or a bow and arrow. Despite the resemblance between the appendage and several familiar objects, at this juncture the intent of its maker has to remain a matter of conjecture. The quadrupeds surrounding the zookeeper are uniform in some ways and variable in others. Measured vertically from back to belly, they range in size from 13 to 18 centimeters. The smaller figures may represent juvenile animals, although —unlike in nature—they have many of the same attributes as the larger ones. As a group, the quadrupeds tend to have elongated, ovoid-to-somewhat-rectangular bodies, straight, thin legs without hooves, and short, straight tails that point upward. The most notable variability occurs in the quadrupeds’ horns or antlers. Some figures have backswept or side-curved horns like mountain sheep, while sinuous lines that bend backward and then forward emerge from the heads of others. A few figures have branching antlers, although none are large or elaborate, and it is assumed that these animals represent deer. A small number of the figures have wide snouts, shorter than usual horn-like appendages, and bodies with slight

Photograph by Lawrence Loendorf.

FIGURE 3.3. A portion of the rock art panel at the Zookeeper site. Note the variation in the quadruped figures and the lines connecting some animals.

90

L AWRENCE L. LOENDORF — THUNDER

AND

HERDS

FIGURE 3.4. A loop-line figure that is connected to several of the quadrupeds at the Zookeeper site. Photograph by Lawrence Loendorf. humps that, taken together, suggest the physiognomy of bison. A number of the figures consist of headless bodies with one leg and sometimes a tail, which led to their original classification as abstract blobs. Since other Developmental Period rock art sites contain similar kinds of figures, it is now thought that they represent partially completed animal figures or perhaps carcasses. The panel contains one solitary, wavy line that may represent a snake, although there is nothing to indicate that the figure is anything other than a wavy line. Many of the quadruped figures are linked by distinctive pecked lines that emerge from the tail of one animal and enter the nose of another. Some of the lines are fairly straight, but sometimes they meander before becoming attached to a companion figure. The rightmost quadruped in the panel is unique in that it is connected not only to another animal but also to a series of nested circles or loops (Figure 3.4).

Placing the Zookeeper Site in Time Several different types of evidence, no one of which is alone definitive but in combination are compelling, make it possible to establish the age of the Zookeeper panel. First is the composition of the panel: many of the figures are connected to one another and have a similar degree of varnish, suggesting contemporaneity. The panel may not have been completed in a single day or by a single artist, but it is almost certainly the product of the same cultural group and was created in a relatively short period of time.

3 — PURGATOIRE PECKED-I PETROGLYPHS

91

FIGURE 3.5. A view of the Point site, located on the prominent sandstone outcrop, as seen from the Zookeeper site. Photograph by Lawrence Loendorf. Second, the analysis of a series of cation-ratio samples from the site produced essentially the same age estimates, supporting the conclusion that the panel’s figures are contemporaneous.10 It is also noteworthy that the cation-ratio dates at the Zookeeper site range from A.D. 900 to 1000 and are consistent with other chronological evidence from a nearby, well-dated habitation locality known as the Point site. The proximity of the Zookeeper site—situated in a small rincon that is part of the Picket Wire Canyon rim—to the Point site, which is located at the terminus of the rim and overlooks the canyon, further supports the assumption that the radiocarbon age of A.D. 920 on charcoal from the latter site closely approximates the age of the Zookeeper panel (Figure 3.5). Finally, the temporal placement of the Zookeeper site has been established through the seriation study presented in Chapter 1, Figure 1.7. The proportion of figure types at the Zookeeper site indicates that the panel was created shortly after depictions of human figures were introduced in the region.

Reflections Stylistically, the Zookeeper site is an excellent example of the Purgatoire Pecked-I rock art site type occurring in the Developmental Period. Although characterized

92

L AWRENCE L. LOENDORF — THUNDER

AND

HERDS

primarily by the presence of animals, this site also contains a significant number of abstract forms. In contrast with sites in the subsequent Diversification Period, animals at Purgatoire Pecked-I sites are not shown with spears or arrows protruding from their backs. The single anthropomorph is very likely an early example of later Purgatoire Pecked-I human-like figures, and its integration into the scene is significant. Several aspects of the Zookeeper site’s setting, in addition to its proximity to the Point site, merit further discussion. The small, 80-square-meter Point site is positioned on a sandstone outcrop that is isolated from the main canyon rim.11 Sheer walls, extending approximately 6 meters above the talus slope, surround the site, necessitating access by now-eroded handholds or by a ladder. Zookeeper, about 100 meters to the northwest, is visible from the Point site, even though juniper trees now partially block part the view. Since vegetation may have been different at the time the two sites were in use, no visual obstruction may have existed. Persons living at the Point site would have repeatedly passed Zookeeper en route from their dwellings to places on the canyon rim where daily tasks were undertaken. There are seven rooms at the Point site, defined by vertically and horizontally laid sandstone slab walls; and the north, west, and south sides of the outcrop are walled (Figure 3.6). An eastern wall also appears to have been present, but at some point in the past it collapsed and fell over the edge of the canyon. The 11-

FIGURE 3.6. The sandstone slab-wall structures at the Point site, which is located a few dozen meters from the Zookeeper rock art panel. Photograph by Lawrence Loendorf.

3 — PURGATOIRE PECKED-I PETROGLYPHS

93

square-meter test excavation into the interior of one room revealed a packed clay floor and an interior hearth. A few fragments of chipped stone debitage and a single rodent femur were recovered from the deposit, and one small, un-notched projectile point or preform, a complete slab metate, a partial metate, and a mano were found on the surface of the site. Based on its size and location, the Point site is thought to have served as a lookout post for a much larger village of between 60 and 70 slab-wall houses, referred to as the Sorenson site. The village is located on a peninsula 1 kilometer north of the Point site, and there is line-of-sight contact between the two sites. Test excavations at the Sorenson site recovered a Scallorn projectile point and flaking debris, while artifacts collected from the site’s surface included cornernotched and stemmed projectile points, bifaces, manos, and metates. A radiocarbon sample from the deposit was dated at A.D. 1020, which is compatible with the time period established for the Point site and the temporal range produced by cation-ratio dating of the Zookeeper site. No ceramics were found at either the Sorenson or the Point site. Because access from the peninsula to the Sorenson site is blocked by several rock walls, it is possible that the location of the village was chosen for defensive reasons. The Point site, on the other hand, could have been successfully attacked by an enemy armed with either bows and arrows or some other type of projectile, such as rocks, which could have been lobbed into the site’s rooms. Vulnerability to attack makes it likely that when enemies were sighted or known to be in the vicinity, the inhabitants of the Point site would have retreated to the better-protected Sorenson site. Another potentially significant factor affecting the relationship of these two sites is the location of the Point site on a major trail leading into Picket Wire Canyon. In the past, this route provided the primary access into and out of the canyon for humans and deer, and presumably also for bighorn sheep. If the inhabitants of the two sites were engaged in communal hunting, the Point site might have served as a lookout for game animals leaving the canyon. In such a scenario, when conditions were appropriate, the inhabitants of the Point site could have signaled hunters at the Sorenson site for help in driving or trapping potential prey. Although conjectural, this supposition is as plausible as the proposition that residents of the Point site acted as observers of the surrounding area, on alert for intruders. One feature of the Zookeeper site may have important implications for why that site was created and how it was integrated into the lives of persons at the two nearby residential sites. I have already noted that the panel is recessed in an alcove whose base forms a shelf or bench along the sandstone wall. Similar shelf features are found at other petroglyph panels in the region, which suggests that the cooccurrence of rock art panels and bedrock shelves is deliberate and that shelves

94

L AWRENCE L. LOENDORF — THUNDER

AND

HERDS

had a function related to the petroglyphs themselves. For example, if a petroglyph panel depicting prey animals served as a prayer station for hunters prior to their departure on a hunt, the shelf may have been where offerings were left. Shared equipment, such as nets used in a group hunt, may also have been stored at such a site, where they would have been accessible to hunters who needed them and knew where they were. One thing is certain, however. Determining the purpose and function of petroglyph panels in the ideational, ceremonial, and practical lives of persons in the past requires the integration of the results of extensive research at the local, regional, and global scales. The Zookeeper site, and others like it, are important to this effort, but time and patience will be required to make such explanations “happen.”

THE BIG HANDS HUNTER SITE The importance of the hand in the mental and physical development of a human infant simply cannot be overemphasized. At birth, each tiny appendage can grasp reflexively any object placed in its palm, even though babies are unaware that they even have hands. Somewhere between six and eight weeks of age, babies accidentally touch one hand with another, and the exploratory world of childhood begins in earnest. At first, because infants are unable to raise their hands to eye level, only their sense of touch is involved, as they repetitively grasp and pull the fingers of one hand with the other. Sometime in the next few weeks, their hands enter their field of vision, and they begin to realize that these fascinating objects belong to them and their movements can be controlled. Fingers can be made to wiggle and wave on command, hands can be put into the mouth, and, as any parent can testify, babies play with their newly discovered “toys” endlessly. In the terminology of the science of infant development, this initial step toward selfhood is called hand-eye coordination.12 The primacy of the hand continues throughout childhood, as children learn to manipulate eating utensils, fasten buttons, tie shoelaces, control toothbrushes and combs, and master a growing number of diverse items such as computer joy sticks, bicycle hand brakes, piano keys, fishing rods, and playing cards. Children are introduced to social rituals like shaking hands, the necessity of raising one’s hand in a classroom, the religious practice of placing the hands together in prayer, and if blind, the process of learning Braille. They learn metaphors such as “hands on” and “hands off,” “handiwork,” “hand-to-mouth,” and “handy.” Parents entertain their children by projecting light onto their hands as they make shadow animals such as bunnies and birds on a wall. Adult human beings seem not to have abandoned the innate fascination with their hands. In the rock art record from the earliest to the most recent of times, the hand is the most common stylistic element. Red hand stencils and prints occur in panels dated between 30,000 and 36,000 years ago at Chauvet cave in France.

3 — PURGATOIRE PECKED-I PETROGLYPHS

95

They occur in large numbers in complex, painted arrangements in caves in Borneo and Argentina, and as single petroglyphs—associated with spirals—at sites at Zuni and La Cienega, New Mexico. Rock art figures resembling humans are sometimes anatomically complete, but often their appendages, including hands, may be disproportionately large, or missing digits, or simply absent. Ethnographic accounts point to some handprints as clan markers or, in other contexts, as symbols of the completion of male and female ceremonies. But the majority of hand imagery in most of its variations remains to be analyzed. This statement applies to the anthropomorphic figure, to be described shortly, whose large appendages provided the inspiration for the name of the Big Hands Hunter site.

Petroglyphs at the Big Hands Hunter Site The site consists of a boulder-strewn area on the north side of the Hogback, within the boundaries of the Piñon Canyon Maneuver Site. On a sloping bench about 60 meters above Van Bremer Arroyo, the site is 35 meters below the crest of the basalt dike. Looking north, the view out onto the steppes is excellent, but there are few properties that would make the site suitable for residential occupation. It is windswept and cold in winter, gnat-infested and hot in summer, there are no nearby permanent water sources, and the vegetation consists of small shrubs and cacti that provide no shade on hot, sunny days. Despite these inhospitable conditions, the presence of 51 petroglyph-bearing basalt boulders attests to the fact that the site had an important place in the lives of prehistoric people and that they spent a considerable amount of time there. The large number of engraved quadruped images suggests that the persons who used the site were focused on game animals and that the site’s primary function was somehow related to them. Many of the quadrupeds appear to have been placed so that they would be easily observed. A good example is the panel containing the figure for which the site is named (Figure 3.7). The boulder itself is approximately 1 meter high, and its flat, upright surface faces north, overlooking the vast, rolling, grass-covered steppes and brush-filled Van Bremer bottomlands. Three principal figures are pecked into the boulder’s surface. One is a large deer measuring approximately 25 centimeters from nose to tail, whose head is topped by a large set of antlers. A long spear protrudes from the back of the deer, which stands behind a full-facing anthropomorph, also 25 centimeters tall, whose outspread arms end in oversize hands. Behind the deer, a 25-centimeter-high series of interconnected circles and loops is stretched like a net. These four design elements—the deer, the net, the spear, and the human figure that has been termed the hunter—are essential components of the huntrelated rock art that appears at numerous places on the Hogback. Similar scenes have been inscribed on many other boulders at the site. The surface of one boulder contains six complete quadrupeds, two of which have

96

L AWRENCE L. LOENDORF — THUNDER

AND

HERDS

FIGURE 3.7. The Big Hands Hunter petroglyph. The panel is on an upright boulder that overlooks the vast grasslands to the north. Tracing by Linda Olson.

branching antlers, one that has looped headgear, and two that have long ears. One quadruped in this group has no head appendages. Two of these animals have spears protruding from their backs, and one of the speared animals has a lighter pecked area around its nose and mouth, which may be meant as an indication that it is bleeding profusely. The panel also includes what appears to be an incomplete quadruped as well as amorphous pecked dots and lines. In this scene, some obviously mature animals are smaller than others and may reflect the artist’s attempt to use diminishing size to indicate that the ungulate herd is fleeing from pursuing hunters. The quadruped that appears to be bleeding from the mouth has an expanding funnel-shaped line connected to its head. On another boulder, a similar funnel-shaped feature occurs on a speared quadruped, from whose head a series of pecked dots floats upward like bubbles (Figure 3.8). Although these lightly pecked elements may represent a physical phenomenon such as blood, they might indicate that the dying animal’s soul or spirit is departing from its body.

3 — PURGATOIRE PECKED-I PETROGLYPHS

97

FIGURE 3.8. One of the bubble-head quadrupeds at the Big Hands Hunter site. Note the figure’s very erect tail. It measures about 25 centimeters from its front to the top of its bubble head. Tracing by Linda Olson. In other examples of the same motif, the dots are less directly associated with the animals but are still near their heads. A half-dozen boulders at the site contain only one or two animals, and these often seem more carefully made than the figures on panels containing multiple quadrupeds. One boulder has a pair of well-crafted deer, one following the other, that are alike except for the first deer’s large, intricate set of branching antlers. The absence of antlers on the second figure suggests the pair is a male and female walking, without upraised tails, in an unhurried way as indicated by their straight legs. A similar pair of figures occurs on another boulder, but in this instance the antlered quadruped is trailing the one without antlers. Large sets of branching antlers are characteristic of many of the quadrupeds at the Big Hands Hunter site. While some figures, such as the buck and doe pairs described previously, have antlers that are relatively realistic, other quadrupeds have such exaggerated and complex antlers that they almost obscure the animal itself. At least one petroglyph artist appears to have become so absorbed in defining a set of

98

L AWRENCE L. LOENDORF — THUNDER

AND

HERDS

FIGURE 3.9. Quadruped petroglyphs from the Big Hands Hunter site. Note how the antlers dominate the head of one figure. Drawing by Chris Alford based on the field sketch by Christine Martinez.

intertwined antlers twisting across the front of a quadruped’s body that depiction of the head was omitted (Figure 3.9). On another boulder, the head of one quadruped supports a massive set of looping and twisting antlers, and a line, possibly representing a short spear, projects from its back. The animal’s rear feet are connected to an ambiguous linear, solidly pecked object, while an amorphous, less densely pecked area near its nose may represent blood. These well-made quadrupeds are very different from a number of partially pecked figures resembling isolated animal body parts whose incomplete state appears to have been deliberate. Images of spears occur on several other boulders, one of which contains a pair of weapons with interesting properties. The grip of one spear is shaped like a bisected circle, and its shaft, forming a right angle at the bottom of the rock, terminates in a quadruped (Figure 3.10). A horizontal line extends from the spear’s midpoint and is attached to another quadruped. The second spear on the same boulder has a three-loop grip, and its shaft extends down the face of the rock. It, too, is intersected by a horizontal line near its midpoint, but the line does not end in a quadruped. What is meant by this less-than-realistic linkage of spears and animals is uncertain, but these symbols appear to represent some aspect of the death of the animals.

3 — PURGATOIRE PECKED-I PETROGLYPHS

99

FIGURE 3.10. Quadruped figures and intersecting spears at the Big Hands Hunter site. Drawing by Chris Alford based on the field sketch by Christine Martinez.

Very few anthropomorphic figures are found at the Big Hands Hunter site. In addition to the eponymous hunter with large hands, there is another full-facing anthropomorph that is approximately 10 centimeters high. This figure is slightly more substantial than the stick figures on nearby sites and has a round head and outstretched arms (Figure 3.11). Although digitate hands are not visible, they are implied by the presence of an S-shaped object at the end of the figure’s left arm. A quadruped behind this anthropomorph looks alarmed, and its posture suggests that it is fleeing. A stylized, less obviously human figure occurs on a panel containing one partial and six complete quadrupeds. The figure is approximately 10 centimeters tall and has a bulbous head and short arms that are attached to its relatively wide body. Only one of the downward-pointing arms ends in shapes suggesting fingers, and the figure’s short legs have no feet. There is an appendage that may represent a penis, although this body part is seldom present on other anthropomorphs in the Purgatoire Pecked-I style. Other less numerous petroglyph elements at the site include depictions of small human feet or footprints that are less than 15 centimeters long (Figure 3.12). One footprint is associated with an anthropomorph, and several are attached to lines connecting them to abstract elements. On one boulder, two figures that may be anthropomorphs are attached—one by appendages that look like legs—to a long, stepped line that crosses the surface. The second figure is connected to a line

100

L AWRENCE L. LOENDORF — THUNDER

AND

HERDS

FIGURE 3.11. Anthropomorph petroglyph from the Big Hands Hunter site. The figure appears to hold something in one hand. Drawing by Chris Alford based on the field sketch by Christine Martinez.

FIGURE 3.12. Human footprints at the Big Hands Hunter site. Human footprints, sometimes in isolated panels, first appear in southeastern Colorado during the Developmental Period. Drawings by Chris Alford based on the field sketches by Christine Martinez.

3 — PURGATOIRE PECKED-I PETROGLYPHS

101

that penetrates an oval loop. A small human footprint on the boulder is connected to the loop and to the anthropomorph. A number of independent linear elements also occur at the site. Some are rectangular forms, and others are simply straight lines—some of which are bisected by shorter lines—that are oriented vertically down the length of the boulders.

Reflections Rock art imagery at the Big Hands Hunter site is dominated by quadrupeds. The 35 complete animal figures constitute 28 percent of the site’s 126 classifiable elements, and if the 20 partial forms thought to represent the bodies of quadrupeds are included in this category, the total number of quadrupeds is increased to 55 individuals, or 44 percent of the site’s rock art inventory. In either case, it is clear that animal figures are a significant component of the total number of images at the site. Based on similarities between the formal properties of the animal images and known characteristics of cervid species, it is likely that most of the quadrupeds are meant to represent deer. As in nature, the petroglyph antlers—when present—are set forward on the head. The presence of upward-pointing tails replicates a distinguishing characteristic of whitetail deer, an abundant species in the region. A single animal figure is shown with a long, low body, a pointed snout, and a long tail. The creature’s general profile suggests that it could be one of several species, from a weasel-like animal to a mountain lion, but it lacks other characteristics that would make identification possible. Its feet, for instance, are depicted with insufficient detail to indicate whether or not claws are present. Mountain lions are, however, the deer’s primary predator, so this anomalous figure may have been intended to suggest the vulnerability of deer to predation by mountain lions. Other aspects of the rock art support the possibility that predatory interactions are depicted in some of the animal representations. The spears protruding from six of the quadrupeds demonstrate a concern with symbolic killing, and the presence of one well-made net and several other possible nets or snares indicates that humans conceptualize deer as prey. At the same time, some quadrupeds are obviously made to look powerful, such as the figures with oversized and unrealistic antlers, and these may represent spiritual figures as much as actual deer. There is an intriguing similarity between the unrealistic antlers on some animals and the images of nets used for hunting. The antlers on two quadrupeds, one on Boulder 16 and the other on Boulder 24, for example, are interconnected in a net-like pattern (Figure 3.13). The juxtaposition suggests that the makers of these petroglyphs were trying to convey their vision of deer as “net heads” as well as “antlered heads.” On Boulder 14, the conflation of antlers and nets is particularly obvious on a figure whose head has been entirely replaced by net-like antlers. Such

102

L AWRENCE L. LOENDORF — THUNDER

AND

HERDS

replacement may relate to the fact that when deer are captured in a net, their antlers invariably become entangled in the woven structure. If the figures on Boulders 14, 16, and 24 are representations of “net-headed” deer, then their symbolic capture is equivalent to the symbolism associated with penetrating spears. In both situations, the animals are depicted as controlled by human hunters. I am comfortable with the conclusion that the Big Hands Hunter site is related to hunting deer, but I am less certain about what it represents. One possibility is that the figures should be viewed as participants in various straightforward hunting tableaux. Evidence against this conclusion, however, includes the fact that there are only five anthropomorphs at the site and they are not in hunting postures. There is just not enough hunt-related interaction—expressed elsewhere as human figures running after prey animals with upraised spears—between the human figures and the site’s many quadrupeds. Although the Big Hands Hunter himself is associated with a net and a speared animal, one would expect more event-oriented scenes, with hunters actually spearing animals and driving game toward a net. Another interpretation of the imagery is that the figures and their physical juxtaposition represent the work of a hunt shaman who used depiction to try to

FIGURE 3.13. Quadruped figure with a net-like head at the Big Hands Hunter site. Drawings by Chris Alford based on the field sketches by Christine Martinez.

3 — PURGATOIRE PECKED-I PETROGLYPHS

103

control the behavior of prospective prey. The net-like antlers and antlers with parallel lines are similar to entoptic forms produced by shamans experiencing altered states of consciousness as they interact with the spirits of game animals during the construal stage of a trance.13 One could argue that secondary evidence supporting this possibility is provided by the figure on Boulder 30 whose spirit appears to be leaving its body.

CONTINUITY AND CHANGE IN PURGATOIRE PECKED-I PETROGLYPHS In this chapter I have discussed the distribution and properties of rock art sites encompassed by the Purgatoire Pecked-I Petroglyph taxon. Although classificatory and temporal units are not always coterminous in archaeology, in this instance the Purgatoire Pecked-I Petroglyph taxon and the Developmental Period coincide on the central High Plains. For almost 1000 years—between A.D. 100 and 1050—sites containing the kinds of rock art figures found at the Bull Pasture, Zookeeper, and Big Hands Hunter sites are the most common in the region. As I noted earlier, the Developmental Period is a time of change, so it is not surprising that there would be corresponding changes in the rock art of the High Plains region. Sites from this period are easily recognizable, for they contain large numbers of quadrupeds whose bodies, legs, and heads have been completely pecked instead of simply outlined. The production technique was not unvarying, however, as exemplified by a much smaller number of quadruped figures that were made by stipple-pecking. The most significant change in the rock art inventory of the central High Plains is the presence of clearly recognizable anthropomorphic figures. Although similar human representations appeared as much as 5000 years earlier in other parts of the intermountain West, including on the northern Plains, and—as I have already discussed—ambiguous, possibly human figures occur occasionally in Archaic rock art panels of this region, there is no doubt about what the new anthropomorphic figures are meant to represent.14 They have fairly substantial bodies with outstretched or sometimes upraised arms and, occasionally, as at the Bull Pasture site, both arms may be depicted as reaching to one side. Their legs are straight, with little indication of bending or movement. The hands of some figures, such as the one from the Big Hands Hunter site, end in fingers, but feet are toeless and usually indicated by L-shaped attachments. Quadrupeds occur at Purgatoire Pecked-I Petroglyph sites in decidedly larger numbers than do anthropomorphs, so it is not surprising that there is more variation in these animal species than in human figures. The cervids with net-like heads and exaggerated antlers at the Big Hands Hunter site, as well as the mix of animals at the Zookeeper site, typify some of the variability. Many figures depict

104

L AWRENCE L. LOENDORF — THUNDER

AND

HERDS

deer and pronghorn, but the large antlered quadrupeds at the Bull Pasture and other sites may represent elk. Quadrupeds with bison-like features also occur in some panels, but they are not common. Distinctive characteristics of many Purgatoire Pecked-I quadrupeds include long, linear legs, whose length is sometimes unrealistically exaggerated. While the legs of some quadrupeds are straight and suggest the animals are stationary, others are positioned to indicate that they are running. Since, in nature, pronghorn are known for their great speed, the depiction of animals in flight is realistic. The feet of individual quadrupeds may be either cloven or L-shaped, or, less frequently, indicated by ball-like attachments. Cloven feet are most commonly depicted by an inverted U-shaped appendage pointing downward from the end of the leg, but a U-shaped foot is sometimes attached to the side of the leg. The term “ball-like” refers to small, circular, solidly pecked blobs attached at the end of one or two— but not always all four—of an animal’s legs. Similarly, L-shaped hooves may occur at the end of one or more of an individual animal’s legs, and in some rock art panels they may be present on one animal but not on its neighbor. The preceding variations in hoof design and placement tend to occur at sites dated to the earlier part of the Developmental Period, but more work needs to be done before variability in this attribute can be relied on as a definitive indicator of Early Developmental sites. Supporting evidence for this generalization comes from the fact that embellishment of the hooves of quadrupeds is less common at sites from the post-Developmental Period. In addition to quadrupeds, the rock art panels at Purgatoire Pecked-I sites contain well-crafted figures that are obviously snakes. In one instance, a rattlesnake is shown striking a deer’s leg, but more often snakes are represented as elongated, undulating figures with triangular heads and pointed tails. I have yet to see any specimens whose tails have well-made rattles. Deer and rattlesnakes have a curiously reciprocal relationship. Not only do rattlesnakes occasionally strike deer in the leg (as depicted in the petroglyph), but deer have been known to take aggressive action in return. I once startled a mule deer in Montana and noticed that it jumped in a strange, stiff-legged way as it ran along the trail. Arriving at the site of the deer’s unusual behavior, I discovered a wounded rattlesnake whose body had been cut by one of the deer’s hooves. Although uncommon, similar deer behavior has been reported by hunters and biologists across the High Plains and may explain why Lakota deer and antelope dancers carry live rattlesnakes in their performances.15 Ungulate hoof prints, human footprints, and bird tracks are important constituents of Purgatoire Pecked-I rock art. Although not present at all sites, they occur with sufficient frequency to serve as identifying characteristics of the tradition. While bird and ungulate tracks occur as far back in time as the Early Archaic Period (5800–3000 B.C.), human tracks are found in the region for the

3 — PURGATOIRE PECKED-I PETROGLYPHS

105

first time at Purgatoire Pecked-I sites. It is unclear whether any of these tracks continue to be made in post-Developmental Period sites. Relatively large numbers of abstract forms—the defining characteristic of the Archaic Period—continue to be present at Purgatoire Pecked-I sites. Pecked lines occur fairly often, as undulating borders that outline panels or, as at the Zookeeper site, as connections between one animal and another. A comparatively large number of quadrupeds have linear, branching designs emanating from their hindquarters. It has been argued, for example, that the lines emerging from an animal’s rear at the Star Tail Quadruped site should be interpreted as a birthing scene.16 The presence at other sites of animal figures with different posterior branching patterns that do not seem to suggest an animal giving birth cast doubt on such an interpretation, however. Circles and circular figures resembling those found in the Archaic Period occur in panels containing animal figures. Some of these, like the group of connected circles at the Zookeeper site, may represent snares, but no correspondence between a variety of other circles—simple, spoked, and rayed—and known material objects has yet to be suggested. The grid-like figures that in some contexts have been referred to as nets are often very realistic. Like quadrupeds, many of them are pecked, although a relatively significant number are made by incising the design into a rock surface. It must be noted, however, that no realistic figures at these sites, such as quadrupeds or anthropomorphs, are made by incising. The placement of net-like forms varies: some occur in front of one or a group of animals, while at other sites they have been placed either beneath or above animal figures. At all sites where incised grid forms are found, however, they are larger than the other figures in the panel. One remaining point in this discussion of linear forms is that parallel and intersecting lines occur at some sites during this time period, but these figures are found in isolation and are clearly not nets or snares. The Purgatoire Pecked-I Petroglyph classification reflects the fact that sites containing rock art belonging to this taxon are concentrated, although not exclusively, along the Purgatoire River and its tributaries in southeastern Colorado. Using Colorado site records as a guide, Purgatoire Pecked-I sites are uncommon north of the Arkansas River and do not extend to the west along the Rocky Mountain front range. In New Mexico, they occur along the Upper Canadian River and its tributaries as far south as Conchas Reservoir.17 A fine example of a Purgatoire Pecked-I Petroglyph site is located on a private ranch near Grenville. To the southwest, still in New Mexico, the figures occur in two or three panels at a major site near Wagon Mound, but they are rare or absent at all rock art sites in the Sangre de Cristo Mountains. To the east and southeast, sites have been found in western Oklahoma, but none, so far, have been reported in western Kansas or western Nebraska. Their absence in these regions may be related to a lack of suitable sandstone.

106

L AWRENCE L. LOENDORF — THUNDER

AND

HERDS

Several distinct patterns emerge when Purgatoire Pecked-I Petroglyph sites are compared with Archaic petroglyph sites. For instance, Purgatoire Pecked-I sites are purposefully separated from nearby Late Archaic sites, as the 200-meter distance between the Big Hands Hunter site and the nearby Hogback Archaic site demonstrates. At the latter locality, basalt boulders whose surfaces would have been suitable for petroglyph pecking were ignored by the creators of the temporally and stylistically different Big Hands Hunter site. This clear spatial separation of temporally successive petroglyph sites is most unusual in other areas of the American West, where sites are often palimpsests of figures from many different groups and time periods. The avoidance of Archaic sites during the Developmental Period is more apparent at localities where petroglyphs were placed on individual boulders than it is in sandstone rockshelters, where the options for placing new images are limited by the available space in the shelter. It should also be noted that the pattern of spatial segregation is not absolute. Figures that were placed on Archaic sites during later time periods, however, most often date to the Protohistoric and Historic eras. In other words, it is more common to find post-Developmental Period petroglyphs or pictographs at Archaic-age sites than Developmental Period petroglyphs. Another characteristic difference between Archaic and Developmental sites involves the proportion of quadrupeds depicted at their sites. At the more recent Big Hands Hunter site, 28 percent of the rock art inventory consists of complete quadruped figures, but if the number of partial or incomplete figures is included in the total, the quadruped component rises to 44 percent. Clearly, many more quadrupeds are present at the Developmental Period site than at the Hogback Archaic site, where quadrupeds make up less than 2 percent of the inventory. Other important patterns help researchers distinguish between early Developmental Period sites and those that precede and follow them. The first regularity involves the frequencies of particular figures at individual sites. At the Hogback 3 site, which is positioned in an exposed location on the north side of the eponymous basalt dike, the distinguishing patterns are easily observable. The site itself consists of a group of boulders that are scattered across a flat bench along the front of the ridge, where it is protected from the wind by another large outcrop of basalt boulders to the south. A single petroglyph depicting a well-made quadruped occurs on an upright, north-facing boulder in the protective outcrop. All other figures have been placed on the boulders scattered across the flat bench. Seventy-three discrete prehistoric elements have been identified on the Hogback 3 boulders, including 11 amorphous pecked forms. The majority of the remaining 62 elements are quadrupeds that can be classified in terms of the appearance of their heads. Twelve animals appear to have ears, horns, or less elaborate head appendages; 11 figures have branching antlers, some of which have multiple branches, and 14 quadrupeds have no identifiable antlers or head appen-

3 — PURGATOIRE PECKED-I PETROGLYPHS

107

dages. In the latter group there are two or three partial figures whose heads are missing. Quadrupeds at the Hogback 3 site, therefore, are distributed nearly equally in each of the three categories of head embellishment—those with ears or horns, those with branching antlers, and those lacking head appendages. None of the appendages are grossly misshapen like those at the Big Hands Hunter site. It is also temporally significant that no human figures occur at the Hogback 3 site, and no spears or arrows are associated with or embedded in the animals. Two rectangular grid forms, one of which is more realistic than the other, appear to be nets. Placement of the Hogback 3 site in time can be estimated by examining the kinds of figures present at the site. The absence of recognizably human figures, which occur at later Developmental Period sites, indicates the site is older than either the Big Hands Hunter or Zookeeper site. A determination of how early the site is in the regional cultural sequence is based on the high percentage of quadrupeds and the presence of figures with branching antlers. These characteristics of the assemblage indicate that the site is more recent than the Late Archaic Piñon Canyon Boulder sites and therefore dates to approximately A.D. 800. Perhaps the most fascinating regularity at the Hogback 3 site is the spatial distribution of petroglyph figures on specific boulders. The two rectangular grid forms identified as nets are located at either end of the site, and all of the boulders on which quadrupeds occur are situated between them. I think this arrangement reflects the fact that deer habitually travel back and forth from the Van Bremer Arroyo bottom to the Hogback ridge, where they pass through breaks in the dike on their way to grazing areas on its south side. On more than one occasion I have seen 10 to 12 mule deer running south out of Van Bremer Arroyo as I approached. They invariably have to run diagonally up the side of the Hogback and through one of the two or three openings in the dike, in their flight passing directly over the Hogback 3 site. It would be difficult for running animals to avoid being caught in nets stretched across one of the openings in the dike, especially in the early morning light, which would make the nets difficult to see. Other features that are characteristic of Purgatoire Pecked-I Petroglyph sites include bedrock metates and abraded surfaces. Sites close to water may contain dozens of bedrock metates as well as large numbers of portable metates. At one site in New Mexico, I counted 27 bedrock metates and another 20 portable metates that had been stacked by the landowner along the fence behind his house. Many of these metates were unusual in that they had deep grinding cavities. It seems possible that the metates had been used to mash chokecherries, buffalo berries, or plums for pemmican or that they had been used for processing acorns. Metates usually far outnumber manos at these sites, which probably reflects the fact that the latter grinding stones were cached in underground pits, a practice that began in the Archaic and continued into the Developmental

108

L AWRENCE L. LOENDORF — THUNDER

AND

HERDS

Period.18 An exception to this pattern occurs at Purgatoire Pecked-I sites along the Piñon Canyon Hogback, where a relatively large number of edge-ground cobbles have been found. These sites also contain significant numbers of boulders with smoothed or abraded surfaces. Recent research by Mark Owens on edge-ground tools suggests they were used for hide processing, an activity that may also account for the presence of boulders with abraded surfaces.19 If animal hides had been stretched over the boulders and then worked with a cobble, it is likely the process would have left its mark on the boulder surfaces.

SUPPORTING EVIDENCE FOR THE AGE OF PURGATOIRE PECKED-I PETROGLYPH STYLE I began this chapter with a discussion of the transitional character of the material culture and rock art of the Developmental Period on the High Plains, and I advanced my belief that the Purgatoire Pecked-I Petroglyph style represents a continuation of Late Archaic pecked rock art. Two sites yielded radiocarbon ages that have helped determine the temporal boundaries of the style. At the Carved Rock site in Picket Wire Canyon, my team and I found a panel of petroglyphs consisting almost exclusively of abstract forms, although it also contained a single quadruped figure.20 In an excavation at what we thought was the base of the panel, we discovered a 30- centimeter below-ground extension of the panel. Charcoal collected from the cultural layer overlying the panel’s base produced a radiocarbon date of A.D. 730, which postdates the manufacture of the petroglyph. Analysis of cation-ratio samples from the panel’s petroglyphs produced a median age of A.D. 150, also suggesting that the rock art panel is older than the radiocarbon age of the charcoal. These two different dates—and the higher ratio of abstract forms to quadruped figures—place the site within the regional transition from the Late Archaic to the Developmental Period, during which quadrupeds were uncommon.21 At the Upper Plum Canyon rockshelter, a radiocarbon date of A.D. 1020 to 1200 was obtained from samples taken from a hearth adjacent to a panel of Purgatoire Pecked-I petroglyphs.22 The date represents an occupation of the rockshelter following the manufacture of the petroglyphs, which according to the stylistic attributes of a quadruped in the panel, should date between A.D. 700 and 800. A reconstruction of the events at the site would proceed, then, from the creation of the rock art panel, followed by the accumulation of a cultural deposit burying part of the panel, to an additional human occupation of the rockshelter during which the hearth was used and then abandoned. According to the chronology that I have developed for Piñon Canyon, quadruped figures became more prevalent during the Developmental Period and, by approximately A.D. 750, groups of quadrupeds, quite possibly representing herds, occurred in the same petroglyph panel. A significant number of

3 — PURGATOIRE PECKED-I PETROGLYPHS

109

abstract forms continued to be represented, and at about A.D. 850, simple anthropomorphic figures with straight bodies and inverted V-shaped legs were added to the Purgatoire Pecked-I Petroglyph inventory. Grids and connected loops that I suspect represent nets and snares also occurred during this period. The Purgatoire Pecked-I Petroglyph style continued into the Diversification Period (A.D. 1050–1450), but the frequencies of some figures changed and new figures were added to the inventory. These changes are significant enough to require the creation of a new stylistic designation that will make it easier to differentiate between older and newer groupings of figures. I discuss this new category of petroglyphs in the next chapter.

NOTES 1 Cole 1985:16. 2 Kalasz et al. (1999:160–88) provide a good summary of the Developmental Period. 3 This statement, made in 2004 by Utah state archaeologist Kevin Jones, was widely reported in

national news outlets.

4 P. Schaafsma (1992:143) includes an image of this panel and suggests that the anthropomorphs

may be holding shields.

5 The Compact Edition of the Oxford English Dictionary 1971: s.v. “hap.” 6 Academy of Achievement 2006:5 (emphasis added). 7 A good overview of the Purgatoire track site is found in Lockley et al. (1997). Information

about the Purgatoire track site is also available at the Comanche National Grassland in southeastern Colorado. 8 Max Canestorp kindly pointed out the inaccuracies in my original account of the events surrounding the discovery of the Zookeeper site and provided me with a replacement scenario. 9 Loendorf and Kuehn 1991:232–49; Loendorf 1992. 10 Loendorf and Kuehn 1991:240–43. 11 Chomko et al. 1990:302. The text of this paper is published in Loendorf et al. 1996:300–313. 12 Spock and Parker 1998. 13 The term entoptics denotes the flickering forms seen in the early stages of altered states of consciousness and was first used by Lewis-Williams and Dowson (1988) in their ground-breaking article arguing for a shamanistic interpretation of rock art. Now called the neuropsychological model, it has been used and published in dozens of places. A good description of the model, with examples, is found in Whitley 2000. 14 Old anthropomorphs are found at Legend Rock, Wyoming (Francis and Loendorf 2002:87, 98) and at sites in the Black Hills (Sundstrom 1990, 2004; Tratebas 1993). 15 Berlo 2000:54. 16 Loendorf and Kuehn 1991:251. 17 P. Schaafsma 1992:144. 18 Loendorf et al. 1996:207–9; Schiavitti et al. 1999. 19 Owens 2006. 20 Loendorf and Kuehn 1991:106–7. 21 A panel at the Carved Rock site also contains anthropomorphs, but these were added during the Protohistoric Period. 22 Quinn 1989:27.

4 PURGATOIRE PECKED-II PETROGLYPHS

common occurrence in science can be summarized as “the more researchers look at a class of phenomena, the more they tend to see.” While sustained looking often reinforces an awareness of the similarities shared by the observed phenomena, what really stands out are the differences among them. On numerous occasions this has occurred in my research at rock art sites at the Piñon Canyon Maneuver Site, in Picket Wire Canyon, and in northeastern New Mexico. In no instance has increased looking produced a greater sense of both similarity and difference than among the figures in the rock art category designated by the term “Purgatoire-style petroglyphs.” In the preceding chapter I described the essential properties of the regional petroglyph style that predominated during the Developmental Period on the central High Plains, whose original designation was simply “Purgatoire Pecked petroglyphs.” But significant changes occurring in this petroglyph tradition during the Diversification Period have prompted me to acknowledge them by adding the roman numerals I and II to the Purgatoire Pecked category. An important distinction between Purgatoire Pecked-I and -II petroglyphs is the temporal correlation between the latter designation and the Apishapa culture of southeastern Colorado.1 Although the prehistoric record of the Diversification Period is complex, most scholars accept that the Apishapa culture developed within the region and that, although not designated as Apishapa during the Developmental Period, groups ancestral to the Apishapa are the most likely source of Purgatoire Pecked-I petroglyphs. In the Ute language, Apishapa is said to mean a “river that stands stagnant.” Why and how long ago the word became attached to the Apishapa River, which

A

110

4 — PURGATOIRE PECKED-II PETROGLYPHS

111

flows east from Colorado’s Sangre de Cristo Mountains to a junction with the Arkansas River near La Junta, is unknown. According to geoarchaeologists, however, the name is appropriate. Soil and pollen studies suggest that, beginning about A.D. 1150, arroyos at Piñon Canyon—and the region’s rivers in general— were flowing less often and with lower volume.2 As in other parts of the American Southwest during the 1200s–1400s, increasingly arid conditions appear to have led regional populations to relocate to more well-watered areas. Because of these demographic shifts, archaeologists are uncertain what tribal groups are represented by the Apishapa phase, although there is a general consensus that Caddoan-speakers such as the Pawnee and Wichita are the most likely descendants of Apishapa-era populations.3 Ethnic identification is complicated by the lack of uniformity in the archaeological assemblages from the Apishapa phase. At some sites, the artifact inventory consists primarily of small, side-notched projectile points and cord-marked ceramics, while deposits at other sites contain predominantly small, cornernotched projectile points and very few ceramics. Inadequate dating of some aceramic sites containing only corner-notched points, which may actually belong to an earlier period, could account for some of the inconsistency, but the presence of assemblages with different artifact types may also indicate that two different ethnic groups inhabited the central High Plains simultaneously. Uniformity is more characteristic of residential structures, however. These were large, composed of multiple units, and could have as many as 40 to 50 rooms—the equivalent of 20 to 30 individual domiciles. Walls were composites of sandstone slabs—many set upright to form the lower part of the structures— and various configurations of leaning wooden poles and brush that formed the upper walls and roofs. These remains indicate that population density in the region, as well as the duration of site use, was increasing. The largest housing clusters are often located in defensible positions on high ridge points, surrounded by precipitous slopes. A series of isolated lookouts surrounded the larger sites, suggesting that inhabitants were either wary of outsiders—although there is no evidence of warfare or raiding—or were monitoring the movement of game animals. Apishapa sites in the Chacuaco region tend to contain more maize and beans than sites farther north. The presence of these cultigens may indicate that better conditions for horticulture existed on the Purgatoire and Chacuaco floodplains, but could also suggest that residents of these eastern sites were more heavily involved in trade with horticulturalists in present-day Kansas and Oklahoma. Evidence of cultigens is scarce at many Apishapa sites, and the remains of extensive seed-collecting and -processing activities suggest that subsistence was focused on gathered wild foods. Large groups of bedrock grinding facilities, sometimes covering as many as 10 square meters, indicate that processing seeds was a communal activity.

112

L AWRENCE L. LOENDORF — THUNDER

AND

HERDS

Faunal remains from Apishapa-phase sites show that populations continued to hunt and consume the same prey species as in the preceding millennium. Deer and pronghorn are the most frequently represented large artiodactyls, but small numbers of bison bones are also present. From the fact that small mammals still significantly outnumber larger body-size species in deposits from this period, it is clear that, as in the past, the meat component of the diet came primarily from small species. Purgatoire Pecked-II Petroglyphs are more dynamic than those of the preceding Developmental Period. Some figures are arranged in narrative groupings, such as those in which small, stick-like human figures chase quadrupeds toward nets or snares. In other scenes where anthropomorphs outnumber quadrupeds, the human figures appear to be dancing or engaged in a social event of some kind. I will illustrate these generalizations in the site descriptions that follow, which have been chosen to illustrate the variation occurring within the Purgatoire Pecked-II petroglyph category.

THE RED -TAIL ROCKSHELTER Where would writers be without bird metaphors? Human virtues and vices are ascribed to birds by authors as diverse as Nathaniel Hawthorne, Edgar Allen Poe, and Agatha Christie, and Shakespeare’s works alone include more than 600 avian references. In American poet Robinson Jeffers’s verses titled “Hurt Hawk,” a redtail is described as strong, arrogant, and dreaming of the freedom of flight. In response to such anthropomorphizing, an ornithologist might be moved to say that red-tail hawks are quite remarkable enough without comparisons to featherless bipeds such as the human species. Possessing eyesight that is eight times more acute than that of humans, they find good nesting places in craggy outcrops with good visibility, where they sit and watch for the small rodents and snakes that make up the majority of their diet. Over a life span of between 10 and 20 years, red-tail hawks help to keep the populations of faster-breeding species in check. Recognition of the extraordinary characteristics of birds in general and hawks in particular was. and continues to be, an important feature of life for historic and contemporary Pawnee peoples, who are thought to be the descendants of the creators of the petroglyph array at the Red-tail rockshelter. Archaeologist Patricia O’Brien has identified several dozen birds that were important in Pawnee cosmology.4 The discovery of an active red-tail nest immediately above the rockshelter—whose official designation is site LA141869—seemed auspicious, so during the recording of the rock art there, and in subsequent reports, we have enjoyed referring to this locality as the Red-tail site. Situated in the high, broken foothills that create the eastern front of the Sangre de Cristo mountain range, the Red-tail site, like many others in the area, is on privately owned New Mexico ranchland. The region, identified as the

4 — PURGATOIRE PECKED-II PETROGLYPHS

113

Raton section, was created by powerful geological forces that caused the earth to uplift, twist, and turn. This produced a terrain that in some places intermixes extinct volcanoes, basalt dikes, and sandstone bedrock, but in other localities forms long, uninterrupted ridges. It is a landscape through which hunters can move undetected in search of game or lie in ambush along known game trails. Communal hunting is rewarded in this terrain, into whose canyons and cul-desacs small herds of animals can be driven into either natural or man-made traps. The rockshelter is situated in the Dakota Sandstone formation known for its high-angle crossbeds like those that tower 12 to 14 meters above the site. Oriented to the north-northwest, its opening is 48 to 50 meters long, and its depth ranges from 2 to 6 meters (Figure 4.1). The sheltered area contains two distinct levels: the floor or current ground level, and a low cavern cut into the sandstone wall about 1.5 meters above the ground level.5 The ceiling height along areas of the ground-level floor is 3 to 4 meters, while the recessed cavern area is 1 to 1.5 meters high. The upper shelf area is 22 meters long and has a maximum depth of about 6 meters.

The Rock Art of Red-tail Rockshelter Five panels of rock art containing several hundred petroglyphs occur along the wall of the rockshelter. The groups of interrelated figures were assigned panel numbers based on their position along the wall’s east-west axis. Quadrupeds,

FIGURE 4.1. An overview of the Red-tail rockshelter. Vegetation in front of the opening obscures the interior of the shelter. Note the red-tail hawk above the shelter. Photograph by Linda Olson.

114

L AWRENCE L. LOENDORF — THUNDER

AND

HERDS

probably representing deer, are the most numerous figures at the site. and although a small number of these are painted, most are pecked. Residual pigment on several of the pecked figures is evidence that they were also painted, although whether completely or minimally is difficult to determine. The quadruped figures, some of which have antlers, are of various sizes. as though the intent was to depict a deer herd made up of adult and juvenile animals. Two large deer, which are 29 centimeters high and 44 centimeters wide, are shown head to head as though their antlers are locked in combat (Figure 4.2). Small human stick figures, some painted and others pecked, surround the deer herds and appear to be driving or chasing them. These anthropomorphs invariably have outstretched arms and wide-set legs but few other distinguishing characteristics. Two anthropomorphs are larger and more substantial than the stick-like figures. One, painted with dark red pigment, is 17.5 centimeters tall and 16 centimeters wide (Figure 4.3). The other figure, which is 12.5 centimeters high and 7 centimeters wide, is pecked. Both figures, each holding a spear or arrow in one hand, have round heads that rest on linear bodies, outstretched arms, and widely separated legs. The held objects have been interpreted by some viewers as atlatls, although for reasons discussed elsewhere in this chapter, I think this interpretation is incorrect. A ladder-like grid or net is a component of several scenes at the Red-tail site. One dramatic, net-like figure—63 centimeters high by 26 centimeters wide— has been placed directly across the path of a group of running quadrupeds that approach it from two directions. In one extraordinary image, several quadrupeds appear to be entangled in a net (Figures 4.4, 4.5).

FIGURE 4.2. Quadrupeds with interlocking horns at the Red-tail site. A portion of a tracing by Linda Olson.

4 — PURGATOIRE PECKED-II PETROGLYPHS

115

FIGURE 4.3. A red-painted anthropomorph at the Red-tail site. Note how the figure is holding an object often identified as an atlatl. A portion of a tracing by Linda Olson.

FIGURE 4.4. A group of quadrupeds at the Red-tail site shown running to and from a net. Small, stick-like anthropomorphs included in the scene are difficult to identify in the illustration. A portion of a tracing by Linda Olson.

116

L AWRENCE L. LOENDORF — THUNDER

AND

HERDS

FIGURE 4.5. Quadrupeds caught in a net at the Red-tail site. A portion of a tracing by Linda Olson.

Many amorphous pecked blobs or areas with no recognizable form occur at the site. In some instances these may represent partially eroded figures, but most are what I refer to as “strike marks.” These are concentrated on the hearts and heads of quadruped figures that appear to have been purposefully hit with a sharp object or perhaps struck by arrows (Figure 4.6). Strike marks are a recognizable component of the Purgatoire Pecked-II and Purgatoire Painted styles and are not unique to the Red-tail site. Very few nonrepresentational figures are included in Purgatoire Pecked-II rock art. Bisected circles, straight lines, and rows of dots do occur, but the style consists primarily of animal and human figures. When the number of figure types at the site is expressed in percentage terms, the results are very similar to element percentages at other Purgatoire Pecked-II sites: 69 percent are quadrupeds, approximately 20 percent are anthropomorphs, and the remaining 5 percent are made up of other figure types. A 13-meter-long area of the shelter wall between the western edge of Panel 3 and the eastern edge of Panel 4 contains no images, which suggests that the placement of rock art figures within the site is deliberate and therefore meaningful. The featureless section of the wall separates the rock art in Panel 3 from a

4 — PURGATOIRE PECKED-II PETROGLYPHS

117

FIGURE 4.6. Strike marks on a painted quadruped at the Red-tail site. Note that the marks occur on the head and in the area of the heart. Photograph by Lawrence Loendorf. concentration of bedrock metates, cupules, and tool grooves associated with Panel 4. Two yellow or gold painted figures have been placed on the rockshelter ceiling above this area. One is a painted starburst design that measures 10 by 11 centimeters, and the other is a 19 by 26–centimeter geometric design or symbol.

The Archaeological Context Heat-cracked and fire-reddened stones are distributed across the floor of the rockshelter, but the most dramatic evidence of the site’s prehistoric use, in addition to the rock art, is the cluster of bedrock metates. A very visible, basin-shaped metate is situated on a sandstone pedestal that is connected to the rockshelter wall at floor level below Panel 2. The metate is approximately 35 centimeters long by 20 centimeters wide, 6 to 9 centimeters deep, and has pigment residue on its margins (Figure 4.7). Ten other bedrock metates measuring approximately 25 by 20 centimeters are found on the floor of the upper level of the rockshelter, which has been designated as Panel 4. Most are shallow grinding slicks, but some are basinshaped and measure 5 to 6 centimeters deep. One metate has a tool groove on its rim, and another has very obvious red pigment near its margins. Six small, round holes ground into the sandstone surface in Panel 4 have been identified as cupules. These features, which are 8 to 10 centimeters in diameter and can be as deep as 8 centimeters, are not stained with the red pigment noted on some metates. A boulder associated with Panel 4 has an extensively abraded area that measures 20 by 33 centimeters.

118

L AWRENCE L. LOENDORF — THUNDER

AND

HERDS

FIGURE 4.7. A deep bedrock metate at the base of a petroglyph panel at the Red-tail site. A residue of red paint occurs around the rim of the metate. Photograph by Lawrence Loendorf. Two broken manos, one sandstone and the other basalt, were noted on the surface of the rockshelter floor. These tools, and undoubtedly many others like them, must have been used in conjunction with the basin-shaped metates, for they are too large for use in the cupules. I do not know what kind of handheld tool would have been used to create the cupules, but I think that some sort of pestle made of stone, hard wood, or perhaps even antler tine could have been used for pounding and grinding. Four sets of tool grooves are also found in association with Panel 4, and another 10 sets of incised lines occur in the Panel 5 area. These grooves developed during the shaping by abrasion of pieces of bone and antler into awls, needles, and handles for scrapers or knives, and they were also used extensively to resharpen bone awls and needles. Some grooves may have been used during the making and sharpening of bone and antler projectile points as well. One set consists of 20 grooves, but most sets contain 3 or 4. A single abraded groove in Panel 5—measuring 34 centimeters long by 1.5 centimeters wide—is larger than the others and hence is somewhat puzzling. No chipped stone tools were found during the rock art recording process. Their absence in part reflects the propensity of visitors coming to see the rock art to look for artifacts on the shelter floor. We did note the presence of small basalt flakes, two quartzite cores, and several small chert retouch flakes from biface thin-

4 — PURGATOIRE PECKED-II PETROGLYPHS

119

ning or resharpening. No ceramic objects or potsherds were found at the site, although the age of the petroglyphs suggests that some should be present in the deposits. No obviously burned bone fragments were noted in the site fill, but these, too, probably occur in buried contexts.

What the Artifacts and Features Reveal about Site Use Because surface materials at the Red-tail rockshelter did not occur in sufficient variety or quantity to justify definitive conclusions about the nature of the site’s use, we decided to explore an eroded channel in the Red-tail drainage, about 60 to 70 meters north of the rockshelter. Sediments in the channel were as much as 1 meter deep in places, and the exposed banks permitted good subsurface examination. No features or artifacts were found in this channel, however, or in other nearby erosional cuts, suggesting that site use was primarily confined to the shelter’s interior. Since extensively used rockshelter sites nearby tend to “spill out” into adjacent areas, the absence of subsurface cultural materials near the Red-tail site may be an indication that it was used as a field hunting camp rather than as a longer-term residential base camp. Based on the presence of classes of artifacts that have specific gender associations, an argument can be made that human use of the rockshelter included family groups. The occurrence on the shelter floor of bedrock metates and broken manos, which are traditionally associated with women’s daily tasks, not only indicates the presence of women at the site but, given women’s responsibility for child care, also suggests that children too were likely to be among the site’s inhabitants. Since rock art scenes depicting deer impaled by spears and other huntingrelated activities suggest a male association with the site, it seems likely that entire family groups were engaged in activities there. Some classes of artifacts may have dual gender associations, however. For example, the bedrock metates were probably used to process seeds and make flour, but they may also have been used to grind the pigments from which paint was made. The well-made metate at the base of Panel 2 is a good example of a multipurpose feature. It is positioned so that it would be difficult to use for dayto-day seed-grinding activities, and its direct association with the rock art suggests a relationship with the pictographs and petroglyphs. On the other hand, even though the residue on the margins of the metate is an indication that it was used for grinding pigment, it may also have been used to process medicinal herbs or some other ceremonial substances that were part of a ritual associated with the rock art. The pigment on the outer edges of one basin-shaped metate associated with Panel 4 supports a similar conclusion. Unfortunately, the upright position of the metate at the base of Panel 2 has made it a popular place for birds to perch, and their droppings have accumulated on the surface, making accurate chemical analysis of the remaining residues difficult.

120

L AWRENCE L. LOENDORF — THUNDER

AND

HERDS

The presence of cupules, such as those on Panel 4, adds an interesting dimension to the discussion of the site’s activities. It is often suggested that such cupules were used for processing piñon nuts, but they actually served a wide variety of other functions. For instance, California ethnographic sources record that small, round holes were made during women’s fertility ceremonies by southern and central tribes and for weather control by northern tribes.6 Cupules may have had a similar function at the Red-tail rockshelter, but there is no corroborating evidence supporting such use. A variety of functions have been proposed for the smaller tool grooves at the site, but as I have pointed out, it is likely that they were used to sharpen bone awls and needles and to rework other bone tools. While the presence of these grooves and other similar features does not provide clear-cut evidence of the gender of persons present at the site, needles and awls were traditionally included in the tool kits of women.7 Because features such as bedrock metates could have been used in the food production activities normally carried out by women or in such typically male activities as grinding pigments or mind-altering substances, the presence of these features at Red-tail rockshelter does not provide unambiguous evidence of the gender composition of the groups using the shelter. Nevertheless, based on another category of data—the spatial distribution of features at the site—I think it is more likely that at least some of the ground stone features were used for food production, mainly because the features are concentrated in an area within the rockshelter where rock art, particularly hunting scenes, is minimally represented. I therefore propose that this area represents the equivalent of kitchen space that was used by women for processing deer carcasses and hides, and that it was a place where women worked independently of men and the male activities associated with the rock art. The yellowish-gold abstract figures on the ceiling above the kitchen area are also important features of this section of the rockshelter (Figure 4.8). No similar designs are found elsewhere at the site, which suggests that the painted images in the metate area are related to the activities that took place there. One possibility is based on the fact that in many hunter-gatherer societies, women were responsible for making clothing, either for personal use or for trade. If the processing of deer and antelope hides into buckskin was one of the tasks carried out in this area, the painted abstract designs may denote the association of the space with such an activity. Alternatively, the images may have been prototypes or patterns that were painted on the processed skins. The latter possibility is supported by the pigment and paint residue remaining in the metates and tool grooves.8

4 — PURGATOIRE PECKED-II PETROGLYPHS

121

FIGURE 4.8. A starburst design painted with yellow pigment on the ceiling of the Red-tail site, directly over the “kitchen area.” Photograph by Lawrence Loendorf.

THE PETROFORM SITE (5LA5555) If numbers mean anything, then the 47,300,000 Google listings for the expression “context is everything” indicate widespread acceptance of the ontological principle that the character of a thing is determined to a significant extent by the circumstances that surround it. This influence of setting on substance applies to words in a sentence, to historical and environmental events, to works of art, to the functioning of bodily organs, and—in archaeology—to the temporal, physical, and spatial relationships between artifacts and features and the deposits in which they occur. Increasingly, archaeologists have enlarged their concept of context to include the ways in which cultural, geographical, and climatic factors have conditioned human behaviors, resulting in distributions of remains at a much larger scale than that of the individual site. In fact, one way of summarizing the changes in thinking in the past 100 years is to say that archaeology has gone from an object-driven to a context-driven discipline. In some instances, however, the vocabulary of archaeology has not kept pace with the increasing precision used by researchers to describe their expanding view

122

L AWRENCE L. LOENDORF — THUNDER

AND

HERDS

of their subject matter. A good example is the elastic term petroform or its anglicized synonym rock shape. These terms, as well as rock-alignments, boulder effigies, and geoglyphs, are used to identify phenomena that can differ greatly in size, shape, design, spatial extent and, presumably, context of origin. The attributes that all petroforms share are few: they are configurations of rocks, and they are the product of human design and construction. An arrangement of small rocks in the shape of a turtle on a bare surface in Whiteshell Provincial Park in Manitoba is called a petroform. So are the interconnected rock features—one of which is more than 30 meters long—at the 9-acre Tie Creek site at the same park. Circular alignments of stone like the Big Horn Medicine Wheel in Wyoming—which may have been used for astronomical orientation—as well as simple but deliberately arranged piles of stones whose function is ambiguous, are all assigned to the same petroform category. These features are likely to remain taxonomic neighbors until more is known about the behavioral context in which petroforms have been created. Numerous circular and linear petroform features have also been documented on the Great Plains, many of which have been identified as medicine wheels. The pattern recognition studies essential to such an effort, however, have been made more difficult by the agricultural activity in eastern and midwestern states which has largely destroyed evidence of prehistoric rock alignments. Commenting on the fragility of petroform features, Jack Steinbring notes that “the mere kicking of a stone may destroy an astronomical alignment. And field clearing, especially on hilltops, may have destroyed hundreds, if not thousands, of such arrangements.”9 Fortunately, time’s ravages have been kind to the petroforms at site 5LA5555 in Piñon Canyon, known unofficially as the Petroform site. Located on a high point of the Hogback, immediately south of the main basalt dike, this concentration of features overlooks Van Bremer Arroyo to the north. The site has a commanding view of the hills and ridges that define the Purgatoire River to the northeast, and the Spanish Peaks and Sangre de Cristo Mountains to the west are in full view. Although vegetation interferes somewhat with visibility in a southerly direction, overall the site provides a panoramic view of the surrounding landscape. The disadvantage of such an unprotected exposure is that the winds common in the area are often forceful enough to make standing upright impossible. The lightning associated with the region’s frequent thunderstorms is also a significant hazard. A number of well-defined petroforms of various sizes and orientations have been constructed at site 5LA5555. They rest on a relatively flat surface whose gravelly base inhibits the growth of ground-level vegetation. Although the Petroform site is not the only location on the Hogback where both rock art and stone alignments co-occur, I am aware of only one other, described subsequently, at which both circular structures and petroglyphs are found. Taking advantage of the propensity of basalt to fracture in flat cleavage planes, the prehistoric inhabitants of the region created walls by placing individual basalt

4 — PURGATOIRE PECKED-II PETROGLYPHS

123

chunks from the adjacent dike in dry-laid layers, one to three blocks high. One petroform feature contains a large 75-centimeter-high block set like a pillar in a semicircular wall made up of stacked layers. At its widest point, this feature measures 2 meters in diameter and has walls that are 20 centimeters high. An abstract petroglyph on the upper surface of the large boulder is made up of circles and curving lines. The largest alignment at the site is a centrally situated, circular feature measuring 4 meters in diameter and composed of 60 to 70 boulders (Figure 4.9). In the remaining intact walls, some of which are 30 centimeters high, the majority of the stones occur in a single layer, although there is evidence that a second course may have been present originally. Another semicircular structure, outlined by 15 to 20 basalt blocks and measuring 2 meters in diameter, is connected to the large alignment on its south side. Some of the remaining walls of the latter feature are 50 centimeters high. No interior features are visible inside either circle. Another interesting feature is the 50-centimeter-high stone cairn that has been placed on top of the dike. At present, it is not possible to say definitively whether the cairn dates to the historic period (although names and dates from that period are found at the site) or whether it is contemporaneous with, and part of, the site’s overall stone alignment complex. No chipped stone debitage or artifacts that might help with chronological placement were recovered from the surface of the site, nor were any bedrock metates or other grinding implements

FIGURE 4.9. The major petroform at site 5LA5555. The site’s petroglyph panels are located on the basalt outcrop immediately above the structure. Photograph by Lawrence Loendorf.

124

L AWRENCE L. LOENDORF — THUNDER

AND

HERDS

present. A reasonable supposition, however, is that the site’s exposure to the elements, the absence of artifacts, and shallow soil development strongly suggest that it was never used as a residence.

Rock Art at the Petroform Site The presence of a dozen petroglyph panels at the Petroform site provides an interesting context in terms of which to view the possible purpose of the rock alignments. Unlike the solitary, static, human-like figure at the Big Hands Hunter site (discussed in Chapter 3), the anthropomorphs depicted on one distinctive panel are unusually action-oriented. Seven figures, shown in profile, occupy the center of the composition, flanked by three quadrupeds (Figure 4.10, left). This positioning reverses the earlier design convention, in which humans were peripheral and quadrupeds were central to a tableau. Exactly what the human figures are doing is ambiguous, but it is clear that they are doing something and that it involves holding hands. One figure—so engaged in the action that it is upside down—faces the other anthropomorphs while holding in one hand a round object similar to those often identified as atlatls. Based on the context, however, I think in this instance the object represents something else, possibly a rattle. With the exception of a single anthropomorphic figure shown in profile and associated with a quadruped, the figures in the remaining panels at 5LA5555 are typical of other petroglyphs in the region. They include fully pecked quadruped figures, ungulate tracks, and static, front-facing anthropomorphs, all of which are characteristic of Diversification Period petroglyphs. Cation-ratio analysis of samples from the action-oriented petroglyph figures produced an age estimate of 700 ± 100 B.P. It is significant that this age is identical with another produced by the same dating technique at a strikingly similar petroglyph site on the Hogback, 8 kilometers to the east. This site, officially

FIGURE 4.10. On the left, dancing figures at the Petroform site, and similar figures on the right from nearby site 5LA5549. The figures at 5LA5549 appear to be dancing. Drawings by Hanna Hinchman based on field sketches by Janet Lever.

4 — PURGATOIRE PECKED-II PETROGLYPHS

125

known as 5LA5549, is also situated on a high point of the basalt dike and has an excellent view in all directions. Several stone circles were originally present, but a road through these features has all but destroyed them. A rock cairn on top of the dike is thought to be prehistoric in origin. There are a number of petroglyphs at 5LA5549, but the panel of interest contains three anthropomorphs shown in profile—their right sides facing the viewer—and two vertical lines or poles (Figure 4.10, right). The human figures appear to be moving in the direction of the poles, and the one closest to the first pole appears to be touching it. The similarities between the action-oriented figures at 5LA5555 and 5LA5549, and the correspondence between the relatively recent cation-ratio dates for both sites, suggest that these two scenes are contemporary and that both fit within the Diversification Period.10 Another set of human figures depicted in profile occurs at a site in Bent Canyon, not too many kilometers to the northeast as the crow flies. One of these anthropomorphs appears to be moving toward an upright, branching straight line that may represent a tree. The site has not been directly dated, but stylistically the petroglyphs resemble the grouped human figures on the Hogback at Site B and the Petroform site. The presence of Apishapa structures at the Bent Canyon site is another indication that these three sites are close to one another in age.

Searching for an Explanatory Context What might the presence of circular stone features in the Piñon Canyon area— set on bedrock or gravelly surfaces and associated with action-oriented petroglyph figures—be telling us? Similar features elsewhere in the western United States have been identified as dance circles, some built specifically as places where representatives of the spirit world could assemble and dance, and others designed for strictly human participation.11 As part of an effort to increase the sources of possibly relevant information about these features, representatives of ten different Indian Nations visited the Petroform site in 2003 and 2004. During this period of consultation, several individuals suggested that the petroglyph panel with action-oriented human figures represented a dance and that the stone alignments were dance circles.12 Another consultant suggested that the abstract petroglyph on the large boulder in one of the rock alignments was meant to replicate the arrangement of boulders on the surface of the site. While the abstract petroglyph does not duplicate exactly the boulder alignments at the site, the suggestion is worth considering because the original structures may have been altered during the seven centuries that have passed since they were first created. Such a proposition also exemplifies the context-driven approach to learning about the past, in this instance through the involvement of contemporary human sources who continue to participate in a tradition of information transmission about the past.

126

L AWRENCE L. LOENDORF — THUNDER

AND

HERDS

THE CORRAL SITE The Corral site is located on an abandoned ranch property known as the John Sanders Cross homestead, which has been recognized by the Historic American Buildings Survey as an excellent example of vernacular architectural style. A petroglyph panel, on the site long before it was homesteaded by Euro-Americans, is found within a historic corral. Numerous images of the property, which is also on the National Register of Historic Places, are included on the website of the Prints and Photographs Division of the Library of Congress.13 The homestead complex includes a bunkhouse, barn, well house, cool house, and chicken coop. The original portion of the main house, which abuts a large sandstone outcrop, was built around 1905 using natural stone and jacal construction. It was enlarged several times between 1910 and 1915, and a final frame addition was completed about 1924. The corral was designed to take advantage of the outcrop’s large sandstone blocks—measuring 6 to 8 meters across by 4 to 5 meters high—which form a three-sided box or rincon. Wooden post-and-pole fencing— a common feature of western ranches—encloses the fourth side of the corral and was used to close gaps between the sandstone blocks. The size and configuration of the corral suggest that it was used to confine cattle rather than sheep. The one prehistoric petroglyph panel at the site occurs on the north side of one of the large sandstone blocks forming three of the corral’s four sides. A post-andpole fence that abuts and partially obscures the rock art panel had to be very carefully removed and replaced in order to record the petroglyphs. Twenty-four quadrupeds are featured in the panel, the largest of which measures 28 centimeters from head to tail (Figure 4.11). Of the 21 figures with horn-like head appendages, probably representing pronghorn antelope, one figure has a split horn. Three quadrupeds lacking head appendages are smaller than the others in the grouping and may represent immature animals. One figure with shorter legs and a long tail is apparently a dog. None of the figures is depicted with hooves of any sort. Seven stick-like anthropomorphs, the tallest of which is 16 centimeters high, are positioned across the top of the panel flanking the quadrupeds. They have straight-line bodies crossed by horizontal lines representing arms, and inverted V-shaped legs. Several elements contribute to the impression that the figures in the panel are components of a scene and are not simply unrelated images. One is a net stretched in a curving arc across the path of the animals. Unlike the loop-lines or snares that occur in some Purgatoire Pecked-I petroglyph panels, the net at the Corral site is so well defined and appropriately placed that its maker’s intent is clear. Another important feature is the well-made figure of a bird that appears to hover at the bottom of the panel near one end of the net. Its vertical body has tail feathers, and the feathers on its outstretched wings point upward. The figure resembles a very

4 — PURGATOIRE PECKED-II PETROGLYPHS

127

FIGURE 4.11. The petroglyph panel at the Corral site. Original field drawing by Linda Olson. similar bird with downward-pointing feathers which is positioned below a group of painted quadrupeds at the Game Drive site (discussed in Chapter 5). At the far left of the panel, a fleeing quadruped with an arrow or spear in its back reinforces the impression that the scene reproduces the figures and activities involved in a communal game drive. Animal figures penetrated by spears occur frequently at Diversification Period sites, so it is not surprising that the age of the Corral site panel has been determined by seriation and by a cation-ratio estimate to date between A.D. 1400 and 1500, which places the site very close in age to the Red-tail site (see Figure 1.7). Both sites apparently date to near the end of the Diversification Period. An important feature of the Corral site is its setting within an archaeological complex remaining from a former pronghorn drive site. Features include a 400-meter-long rock alignment of tabular sandstone slabs that follows the southern rim of Lockwood Arroyo (Figure 4.12). The alignment terminates in a concentration of collapsed rock at its eastern end. I believe that the tabular rocks were originally set upright to form inverted V-shaped bases which supported flat, horizontally placed stones, creating structures that would have looked somewhat like small human figures with outstretched arms. Although no excavation in the area of the drive line has yet occurred, I suspect that the now-jumbled concentration of rocks may once have formed a “calling station” where, hidden from view, a hunt leader used his voice to coax antelope into the upper reaches of Lockwood Arroyo to the northwest of the drive

128

L AWRENCE L. LOENDORF — THUNDER

AND

HERDS

FIGURE 4.12. A map showing the relationship of the features at the Corral site. Illustration by Bonita Newman.

line. Once the animals were surrounded on two sides by the banks of the arroyo, a group of hunters would have chased them eastward until they were trapped in an area between the sandstone blocks to the north and the drive line with its human effigies to the south.

4 — PURGATOIRE PECKED-II PETROGLYPHS

129

Historical accounts of pronghorn drives indicate that at this point the confined animals would have been forced to run in a circle until they collapsed. By stationing a line of human hunters across Lockwood Arroyo to the east, whose dog travois poles might have been set on end to form a fence, the animals could have been driven toward the sandstone outcrop that contains the rock art panel.

Communal Pronghorn Hunting The archetypal communal hunting sites on the Plains, particularly the famous buffalo jump sites in Alberta, Montana, and Wyoming, have been recognizable archaeologically because so many large animals died during the game drives that hunters were unable to butcher all of the carcasses and left many behind untouched. The smaller body size and different behavioral characteristics of pronghorn antelope— which the rock art figures at the Corral site seem to depict—as well as the nature of the communal hunting process itself account for the fact that archaeological sites containing unexploited pronghorn carcasses are rare on the High Plains. I am not aware of any recorded sites containing the remains of large numbers of deer carcasses, because this species does not assemble in large enough herds for there to be any unprocessed carcasses remaining after butchery. If, for example, a group of hunters had chased a herd of 10 to 15 deer and successfully captured 6 or 8 of them, they would have been able to butcher all the carcasses and transport all the meat, leaving behind very little evidence of their activities. For the same reason, the bones remaining from a communal pronghorn hunt would be almost undetectable. Nevertheless, my team and I decided to place a series of auger probes into the sediments at the Corral site to see if we might locate any evidence of a prehistoric antelope-processing area. A low sheltered area in the rocks on the south side of the large boulder containing the petroglyph panel was selected for two auger probes because flaking debris and heat-cracked stones were visible on the surface. Although one chert flake was recovered in one auger probe, there were no bone fragments, charcoal, or any other evidence that a communal pronghorn drive had taken place at the locality. These results do not, however, negate the possibility that the processing of antelope carcasses occurred there, since the heavy use of the corral in the last century by cattle and sheep would have destroyed the most important evidence. Three auger probes—one placed in the loamy soil immediately below the petroglyph panel and two others placed in a line in the clay sediments of the corral—produced more informative results. All three were dug to a depth of 70 centimeters or more (Figure 4.13). The two probes farthest from the panel contained only clay, with few or no rocks and no cultural remains. In the probe immediately below the panel, however, we discovered a layer of fragmented red and yellow ocher at a depth of 15 to 18 centimeters. These friable, brightly colored fragments

130

L AWRENCE L. LOENDORF — THUNDER

AND

HERDS

FIGURE 4.13. Archaeologist Mark Owens with the auger probe at the base of the Corral site panel. Photograph by Lawrence Loendorf. would have been an excellent source of pigment which could have been used to enhance the petroglyphs, or for some other purpose associated with the rock art. Their location at the base of the panel corresponds to the place where one or more individuals would have stood to make the figures in the panel or to interact with the panel at some point after it was made. Based on our discovery of pigments associated with the pecked quadrupeds at the Red-tail site, we immediately suspected that the red and yellow sandstone fragments at the Corral site might have a similar association. Careful inspection of the petroglyphs did not reveal any evidence of pigments, but to determine if any pigment residue invisible to the naked eye remained on the petroglyphs, we used a portable x-ray fluorescence instrument to run a series of x-ray probes of the panel and the surrounding sandstone surface. We learned that iron levels on the surface of the panel were several magnitudes higher than on the surrounding sandstone surfaces. We also found a high level of strontium on the panel surface and in the red pigment source, but not on the nearby sandstone. The most fascinating aspect of the study was the indication that red pigment had been smeared across the entire surface of the panel and not just on individual quadruped figures. I have no explanation why the application of pigment would have been so extensive, but perhaps it may have been intended to darken the panel’s pale sandstone background and provide a greater contrast with the light-colored petroglyph images. Even if this supposition is incorrect and other

4 — PURGATOIRE PECKED-II PETROGLYPHS

131

unknown factors are involved, x-ray fluorescence analysis did provide evidence of the application of the pigment to the petroglyph panel. Although this discovery does not tell us anything about whether the site was used as a game drive, it does suggest that some ritual took place at the petroglyph panel. The buried pigment may also be shown to be important, if excavation can uncover it in a context with charcoal to obtain a radiocarbon date.

ROBUST ANALYTICAL ASSERTIONS AND SHADOWY APISHAPA Recent observations and analyses of the Developmental and Diversification Period rock art on the central High Plains support a number of statements about the relationships between the categories of pecked figures that I have discussed in this and the previous chapter. I can say with confidence, based on similarities in manufacturing technique and in the formal properties of quadruped and anthropomorphic figures, that the Purgatoire Pecked-II petroglyph style is an outgrowth of Purgatoire-I style. Furthermore, the development of Purgatoire Pecked-II Petroglyphs from earlier forms is consistent with regional archaeological interpretations of Apishapa sites as an ongoing cultural tradition with antecedents in the Developmental Period. It is also clear that the Purgatoire Painted figures discussed in Chapter 5 are contemporaneous with Purgatoire Pecked-II petroglyphs, a conclusion demonstrated at sites such as Red-tail rockshelter, where painted and pecked figures are intermixed with and superimposed upon one another and where some figures were pecked and then painted. I therefore include both Purgatoire Pecked-II Petroglyphs and Purgatoire Painted figures in the term Purgatoire II rock art. When compared with other regional styles, one of the most distinctive characteristics of Purgatoire II rock art is the presence of a large number of anthropomorphic figures. These range from simple, stick-like characters to more substantial, front-facing human forms, to groups of anthropomorphs shown in profile view. The more fully defined figures are shown with hands and feet but lack facial features such as eyes or a mouth and rarely wear headgear. The figures are almost always solidly pecked rather than outlined and occur in sets of two or more (Figures 4.14, 4.15). If isolated anthropomorphs appear on a single boulder, there will invariably be other human figures nearby. Stick-like figures appear to be chasing animals toward nets, while anthropomorphs in profile view are arranged in scenes— sometimes with other human figures—and in postures suggesting that they are dancing. In some rock art panels they occur with both human and animal figures. Species diversity increases in Purgatoire II rock art. Deer and antelope continue to predominate, but other recognizable mammals include bears and dogs or coyotes (Figure 4.16). Dogs are associated with other animals in scenes like

132

L AWRENCE L. LOENDORF — THUNDER

AND

HERDS

FIGURE 4.14. A single boulder at site 5LA5589. Note that the anthropomorphs on the side of the boulder appear to spear the quadrupeds on the top of the boulder. Illustration by Chris Alford from a field sketch by Christine Martinez.

FIGURE 4.15. A single boulder in the Purgatoire River bottom, with a hunting scene on its sides and top. The scene is very similar to the boulder at site 5LA5589. Illustration by Chris Alford from a field sketch by Daphne Rudolph.

4 — PURGATOIRE PECKED-II PETROGLYPHS

133

the one at the Corral site, where they appear to be part of the chase or hunt. A few animal images resemble bison, although realistic bison figures do not appear regionally until the Protohistoric Period. A limited number of object types are represented in Purgatoire Pecked-II rock art. I have discussed scenes featuring obvious nets or fences that have been placed in the path of running quadrupeds. Some animal figures, like the one at the Corral site, have spears or arrows in their backs, and while there are rare instances in which human figures are shown gripping a bow (Figure 4.17),

FIGURE 4.16. A bear and an unidentified companion figure at site 5LA5589. The site is dated to the Diversification Period. Illustration by Chris Alford from a field sketch by Christine Martinez.

FIGURE 4.17. A very rare example of a Diversification Period hunter with a bow at site 5LA296. The figure is about 8 centimeters tall. Drawing by Chris Alford from a field sketch by Daphne Rudolph.

134

L AWRENCE L. LOENDORF — THUNDER

AND

HERDS

images of bows and arrows unassociated with either humans or prey animals are unknown in this time period. The most common object shown in the hand of an anthropomorph resembles a spear or arrow, although some researchers have identified these as atlatls. I am unconvinced by these assertions because of the contrast between what is known about the use of hunting gear in real life situations and several details of the scenes in which these objects occur. In more than one rock art panel, a human figure is shown holding the circular end of a supposed atlatl at the same time that a quadruped is shown pierced by the opposite end of the object (Figure 4.18). Such a connection between a hunter, an atlatl, and a prey animal would never occur in the real world. In order to make effective use of an atlatl, it must be held in an upright position; yet in every rock art representation of an anthropomorph holding one of the supposed atlatls, the object is pointing downward. Moreover, it is relevant to note that thrusting spears, which are used to dispatch animals that have already been incapacitated, would have been held pointing downward. To my knowledge, no thrusting spears have been recovered from archaeological sites in the region, so whether they had finger-loops, as the rock art depictions suggest, is not known.

FIGURE 4.18. One of several examples of an anthropomorph at site 5LA5589 whose hand is connected to a spear embedded in a quadruped. Drawing by Chris Alford from a field sketch by Christine Martinez.

4 — PURGATOIRE PECKED-II PETROGLYPHS

135

The three sites discussed in this chapter are representative of the types of variability occurring in Purgatoire II rock art sites. At Red-tail rockshelter, and at several other regional sites that I haven’t discussed, scenes of humans chasing animals toward nets occur. The physical setting and archaeological contexts of these sites— unlike the Corral and Game Drive sites—argue against the occurrence of activities associated with a communal game drive. One is left with the conclusion that similarities in rock art imagery but differences in physical context seem to suggest that activities at the two types of sites differed in some significant but as yet unknown way. An important aspect of Purgatoire II rock art sites is the prevalence of a large number of concavities or strike marks on panel surfaces. Some researchers have argued that bits of paint or sandstone were deliberately removed from rock art images and placed in the medicine bundles of shamans. While I agree that painted images in the Four Corners region of the American Southwest have been modified in this way, some other activity must account for the strike marks associated with Purgatoire II rock art images. The gouges are so deep at the Red-tail site, for example, that they look less like the results of a controlled process to remove paint and more like impact fractures. The depth and extent of the marks on some figures at first suggest that they are bullet holes, but closer inspection of the rock surface indicates that the source of the depressions lies in prehistory rather than in the fairly recent past. A possible explanation of why Apishapa hunters would have struck the painted and pecked quadrupeds at Purgatoire II sites can be found in ethnographic sources. There the practice of shooting arrows at rock art images of animals, or striking them with a lance or spear, is described as a process intended to magically kill prey in order to ensure the success of an upcoming hunt. A twentieth-century instance of such a practice was told to anthropologist Paul Radin by Crashing Thunder, a Wisconsin Winnebago Indian. Crashing Thunder’s father controlled 20 spirits who lived in a cave and were referred to as “Those-who-cry-like-babies.” When [Crashing Thunder’s father] wished to bless a man he would take his bow and arrows and, holding them in his hands, lead the man around the hill and into the lodge [the cave or home of the spirits]. There he would look for a stone pillar, and upon it, at about arm’s length, he drew pictures of a number of different animals. My father possessed only one arrow, but that was a holy one. Then dancing around the stone pillar and singing some songs, he finished by breathing upon the pillar. Finally he walked around it and shot at it and when he looked at the stone it had turned into a deer with large horns which fell dead at his feet.14 Belief in the efficacy of striking rock art images has been documented in geographically distant cultures. Matilda Stevenson described how hunters at Zuni Pueblo in New Mexico sought magical intervention to ensure a successful hunt by shooting arrows at the petroglyph of a deer.15

136

L AWRENCE L. LOENDORF — THUNDER

AND

HERDS

Arrows were found in a crevice associated with rock art in Arrow Canyon near Nevada’s Moapa River, and another similar set of arrows shot into the rock was associated with a petroform near Borrego Springs, California.16 In a similar account, hunters belonging to the Northwest Coast Lillooet tribe were reported to shoot with miniature bows and arrows at grass or bark replicas of prey animals while on vision quests. An individual would pray “that he might become an expert archer. If he made many hits, he would become a successful hunter. Afterwards he suspended the figures, with the arrows in them, from the branch of some fir tree.”17 Further support for this practice is provided by Thor and Julie Conway, who discovered a historical account from A.D. 1800 by Daniel Harmon describing Ojibwa shooting arrows at the Agawa pictograph site on Lake Superior.18 They indicate that firing arrows at rock art sites was widespread in the region, noting that modern evidence for this practice is evident at some sites. A pictograph cave in the upper peninsula of Michigan has been studied by archaeologists. By digging into the cave floor below the paintings, the archaeologists found the shattered tips of flint arrows or spear points. Nearby, the broken point bases were uncovered from the same weapons. After excavation, the bases were refitted to the tips.19 Although archaeological evidence and ethnographic analogy provide us with a model of the behaviors and beliefs that might account for the presence of strike marks, in most rock art settings archaeologists are unable to determine the ethnic identity of the persons doing the overstriking. At the Red-tail site, however, careful examination revealed the presence of pecked quadruped figures with gouges that were later painted over with typical Apishapa figures. The sequence leaves little doubt that during the Apishapa period, sharp implements were used to strike rock art quadrupeds, and that painted quadrupeds were later placed over some of those images. Evidence of strike marks at the Corral and Game Drive sites is minimal, but pigment appears to have been added to figures in the Corral site hunting scene, suggesting that sites with hunting scenes were used repeatedly. Overpainting makes even more intriguing the question of what High Plains hunters were accomplishing by putting hunting scenes on rocks. Researchers have formulated two principal explanations for the presence of rock art panels with imagery suggesting communal hunting activities. The first proposes that the panels were created by shamans who used the figures of animals in activities meant to control the outcome of hunting endeavors. An alternative explanation proposes that the scenes were created and used by persons responsible for teaching young hunters how to participate effectively in communal hunting situations.20 The imitative magic explanation argues that rock art figures of humans and animals were produced by shamans who imbued the images with power acquired

4 — PURGATOIRE PECKED-II PETROGLYPHS

137

in trance-like states. The powerful images would have been created prior to a hunt, and hunters may have revisited the site numerous times to absorb more of the power concentrated there. The instructional explanation asserts, in contrast, that experienced hunters would have used a dynamic scene of human and animal figures interacting in a rock art panel as a mnemonic device to teach young, inexperienced males the techniques of controlling and killing game. James Keyser and David Whitley have recently discussed how imitative magic, practiced in a hunting context, might be reflected in rock art imagery, and I expand on this idea in the final chapter of this book.21 Efforts to determine whether either of these alternatives explains the High Plains rock art hunting scenes have relied on the regional ethnographic record. In Pawnee ethnography, for example, an account with great time-depth provides an explanatory context for the direct association of bird petroglyphs with the human and animal figures at the Corral and Game Drive sites, and at other sites in the region as well. Sources recount the experience of a young man who had highly developed hunting skills that he often flaunted by killing animals and then leaving them to decompose. Attempting to stop such wasteful behavior, the buffalo sent four female calves disguised as beautiful young women to coax the young hunter out onto the Plains, where he was captured and placed at the top of a high mountain located in the middle of a large lake. The captive went without food and water for several days, during which time he cried and begged for the power to free himself. Eventually, birds of all kinds answered his pleas and helped him to escape. As Pawnee sources explain, “The birds had given the boy their power, including the ability to transform himself into a bird, and the buffalo were finally defeated.”22 Although buffalo are central figures in this story, Caddoan peoples regularly featured both deer and buffalo in their narratives. There are even accounts of competition between deer and buffalo.23 In some instances the deer drive away the buffalo, and in others the buffalo win the right to be with the people. These variations probably correspond to periods in which buffalo populations on the Plains fluctuated between abundant and scarce, but the tale is as effective when deer or antelope are featured or when buffalo play the central role. One would, in fact, expect deer and antelope to have been represented more frequently in Apishapa tales because the archaeological sites produced by their activities—reflecting environmental realities—contained far more deer and antelope remains than those of bison. Given such a world view, it would have been normal for an Apishapa shaman to transform himself into a bird during a vision and then represent himself in avian form in deer and antelope hunting imagery. Before this interpretation can be accepted, however, it is necessary to ask whether the presence of birds in rock art hunting scenes proves that the images were produced by shamans. The answer is “not necessarily,” because if the panels were created by older hunters who were

138

L AWRENCE L. LOENDORF — THUNDER

AND

HERDS

teaching younger men to hunt effectively, the tale describing the fate of the captured, wasteful hunter might have been known to the instructors, who could have recited it and inscribed a bird image to reinforce the story’s lesson. Similar ambiguity surrounds many other rock art images and features. While the animals depicted with arrows in their backs, the nets they were being driven toward, the blood flowing from their noses, and other representations of dying animals may indeed represent a shaman’s attempt to ensure the killing or capturing of animals using imitative magic, they may also represent important technical aspects of hunting that a teacher would want to explain to a pupil. Similarly, actions such as adding pigment to a panel containing a hunting scene or striking the figure of a deer with an arrow could also be explained in a pedagogical context, as ancillary but essential procedures that a novice hunter would need to understand and perform. So far in this discussion, it has not been possible to confirm or reject either of the competing interpretations of the purpose underlying the rock art depiction of hunting-related imagery. Each is plausible but neither is definitive. Now I want to introduce a category of evidence—based on the depiction of unrealistic antlers on some quadrupeds—that I believe supports the association of shamanism with rock art animal figures. The antlers of a number of quadruped figures resemble nets as much as they do antlers and, on some figures, the netlike antlers have replaced the figure’s head entirely. Since, in reality, the antlers of deer captured by net hunting are invariably entangled in those nets, it is easy to appreciate why antlers and nets might have become combined in a hunter’s mind and substituted for each other in an instructional rock art panel. If, however, the unrealistic antlered figures are in fact “net-headed” deer that have been placed under shamanic control during visionary trance experiences, then their symbolic capture is being represented for the benefit of human hunters who then draw power from the images in the same way they do from animal figures penetrated by spears. While zoologists might point out that some cervid species—particularly whitetail deer—develop atypical antlers, these deviations from the normal do not include the bubble-like or candelabra-like shapes that occur in rock art panels. Interestingly, these and other forms such as nested curlicues and connected circles that replace anatomically correct antlers in some rock art figures are identical to recognized categories of entoptic phenomena that can occur during altered states of consciousness. This correspondence is another indication that unnaturalistic quadruped figures were derived from shamanic trance-like states. Evidence that more recent Caddoan shamans were engaged in hunt-related trance experiences specifically related to elk and deer is found in a narrative told to ethnographer George Dorsey by a Skidi Pawnee man named Yellowbird. The story describes a man whose wife had abandoned him and their son. Leaving his

4 — PURGATOIRE PECKED-II PETROGLYPHS

139

village with the boy, the man sought shelter in a deep cave with “drawings of people and animals” on its walls.24 For reasons that become apparent later, the man decided that the cave was unsatisfactory and moved with his son to a smaller cave, which he first searched carefully for the presence of animal tracks. The father left the boy alone in the cave when he went out to hunt. During one of the father’s absences, from the depths of the cave the boy heard the sound of animals singing and dancing. Following the sound, the boy joined the singing animals, learned about elk medicine, and returned to his father dressed in a white skin and wearing red berries around his neck. On a visit to his father’s village, the boy taught the Pawnee community about the magical powers he had learned from the elk. Later, because the father had inadvertently placed his son in a setting in which he had contact with animal spirits, he lamented the outcome, which he regarded as a mistake.25 Several parts of this story have an interesting connection to rock art. First, the father carefully inspected both caves for signs of animals, and he avoided staying in the first one because of the drawings of animals on its walls. Such behavior would not have been unusual, because of the widespread belief on the northern Plains that animals come from caves or from supernatural homes under the ground.26 More importantly, there is the strong implication that the drawings represented animal spirits living in caves. The red berries that the young boy wore as part of his new costume are also important details. In his book on Caddo lifeways, Dorsey described the Wichita deer dance (called the “Medicine Men dance” by the Skidi Pawnee) in which red mescal beans were used to induce trance states: The dance was held generally three, occasionally four, times a year: the first occasion when the grass had just appeared, the second when the corn was ripe, the third when the corn was harvested. The ceremony was never held in winter. One of the special features of the ceremony was the administering to the novitiate of a small red bean, which produced a violent spasm, and finally unconsciousness, this condition being indicated by the inability of the novitiate to suffer pain when the jaw of a gar-pike was drawn over his naked body.27 I suspect that similar dances with animal themes are depicted at the Petroform site, but whether they are related to the Caddoan deer dance has not yet been determined. If sites with relevant rock art imagery contained more detail, we might be able to find “red berry” necklaces, but until other dance sites are found containing the level of detail of the Bear Dance site, which is discussed in the next chapter, this suggestion remains speculative. The presence and activities of deer shamans among the Skidi Pawnee and the Wichita are well documented. According to Dorsey, a Skidi Pawnee man named Scabby-Bull had powerful deer medicine that allowed him to transform

140

L AWRENCE L. LOENDORF — THUNDER

AND

HERDS

himself into a deer.28 Scabby-Bull once demonstrated that he could magically imprint pictures of the moon and stars on plum pits, an act suggesting he would also have been able to make pictures appear on rocks. He also took a friend to a circular wooded area to demonstrate his medicine, which involved running around and around in a circle in much the same way that a deer would have behaved if penned in on all sides by a deer surround. Of course, neither the stories compiled by folklorists nor the documentary record of early twentieth-century deer medicine is proof that Apishapa-period rock art was created by shamans and placed on rock panels during or after their out-of-body experiences. It is clear, though, that rock art played an important role in hunting deer and pronghorn antelope. Given the descriptions of deer ceremonialism in the ethnographic record of the Caddo, Pawnee, and Wichita descendants of Apishapa peoples, there is a possibility that deer dances or similar ceremonies may have taken place at rock art sites in the past and in the near present as well. It is also possible that Purgatoire II rock art sites served as shrines or prayer stations visited by hunters before they undertook a hunt. These possibilities are not mutually exclusive, and there is a good chance the sites were made by shamans and later used by hunters as shrines. More recent ethnographic accounts of pre-hunt rituals to ensure success provide details of a Skidi Pawnee ritual associated with a circle or semicircle of buffalo skulls.29 A similar circular feature made of buffalo skulls was depicted by the nineteenth-century artist Alfred J. Miller.30 An informant whose name was Good Heart told Pawnee scholar James Murie a story about a hunter who found a circle of buffalo skulls that had a stone sculpture of a buffalo in the center.31 The hunter lit his pipe and blew smoke into the nostrils of the buffalo figure, praying at the same time for help in finding his intended prey. Continuing to hunt and pray, the hunter looked out on the Plains and saw a herd of buffalo. Filled with joy, he ran back to the stone buffalo and again offered prayers. Frank North, the Nebraskan who commanded a Pawnee scout unit in the late 1800s, also witnessed a ceremony in which hunters prayed to a circle of buffalo skulls.32 Although the prey animals in the preceding accounts are buffalo rather than deer, the described behaviors are relevant to the present discussion because they reflect the most recent examples of a cultural tradition of reverence and ritual among hunters belonging to a thousand-year temporal continuum of related peoples. Apishapa sites on the High Plains containing rock art images of deer and antelope—but few bison—appear to be near the point of origin of such a tradition and, if the central High Plains had been a good habitat for bison during the Late Prehistoric Period, it is likely that these animals would also have been found in the hunting-related rock art panels. These accounts, as well as the strike marks and added pigment in High Plains rock art panels containing communal hunting scenes, strongly suggest that the rock art occurs in locations that

4 — PURGATOIRE PECKED-II PETROGLYPHS

141

were used as prayer stations by shamans who maintained a connection to the spiritual domain which was expressed in powerful imagery for the benefit of persons responsible for maintaining the physical well-being of the group.

NOTES 1 Kalasz et al. (1999:189–221) present an overview of Apishapa archaeological remains. 2 McFaul 1991:38. 3 Echo-Hawk (2000) presents an excellent summary of Apishapa archaeology as it relates to

Caddoan oral traditions.

4 O’Brien 1986. 5 Individuals seeking more detailed information about archaeological sites in New Mexico

should consult the New Mexico Laboratory of Anthropology website for maps and floor plans (www.miaclab.org/collections/index.html). 6 See Whitley 2000:98–100 for ethnographic accounts of the making of cupules in California. 7 See Sundstrom 2002 for more discussion on tool grooves and female tool kits. 8 These abstract painted designs associated with bedrock metates are discussed again in Chapter 5. 9 Steinbring 2003:2; Steinbring et al. 2003. 10 Loendorf 1989:268, 345–46. 11 Johnson 1986. 12 Blythe 2007. 13 Images of the John Sanders Cross homestead are available online at http://lcweb2.loc.gov/pp/ hhhtml/hhTitles181.html. 14 Radin 1916:27–28. 15 Stevenson 1904:439. 16 Both accounts are from the Lost and Found section of the Journal of California and Great Basin Anthropology (27[1]:79–81). The Arrow Canyon report is an anonymous piece from the Los Angeles Times, October 5, 1904; the Anza-Borrego account is from Begole 1985. A third account describes arrows in a high crevice, but there is no mention of associated rock art. 17 Teit 1906:266 18 Conway and Conway 1990. 19 Conway and Conway 1990:49 20 Sundstrom 1989:160; 1990:282–83. 21 Keyser and Whitley 2006. 22 Harrod 2000:73. 23 Harrod 2000:66–68. 24 Dorsey 1904a:195. 25 Dorsey 1904a:194–99. 26 The belief that animals live in caves or beneath the ground is widespread in the Plains region. This subject is discussed in many sources, including two recent ones. Patricia Albers (2003) presents Lakota accounts of bison that are associated with Wind Cave, South Dakota, and Peter Nabokov (2006) compiles information about Pawnee animal lodges. See also Sundstrom 2006. 27 Dorsey 1904b:16. 28 Dorsey 1904a:231–38. 29 Grinnell 1889:270. 30 Miller 1973. 31 Murie 1984:443. 32 Wilson 1984:47.

5 P U RG ATO I R E PAINTED ROCK ART OF THE HIGH PLAINS

heatergoers in the sixteenth century are said to have greatly admired the spirited, sometimes racy plays of dramatist John Heywood. His concerns weren’t lofty. In productions of “The Mery Play betwene Johan the Husbande, Tyb the Wyfe, and Syr Jhan the Preest” and other comedies, he told tales of everyday people, some of whose colloquial expressions he later assembled and published in a manuscript titled “Proverbs.” Many of the entries are still familiar in the twenty-first century. Who has not expressed—in slightly modified form— the realities of everyday life in the following terms: “haste maketh waste,” “look ere ye leap,” “would ye both eat your cake and have your cake?”1 For rock art researchers, the definition of a class or category of images usually develops from systematic looking, or as Heywood’s maxim puts it: “Proofe upon practise, must take hold more sure/Then any reasoning by gesse can procure.” Some intense looking and practicing was involved during the definition of the Purgatoire Painted rock art style by Sally Cole.2 Analyzing painted images from 11 sites at the Piñon Canyon Maneuver Site and from several sites in the surrounding region, Cole systematically, and with as little “gessing” as possible, noted sufficient similarities among the figures to be able to develop a stylistic definition. She also observed the commonalities between the newly established Purgatoire Painted style and the Purgatoire Pecked rock art in the region. There are 150 Purgatoire Pecked sites and only 25 Purgatoire Painted sites that have so far been identified in the region, making it clear that painted sites are comparatively rare. Since nearly all Purgatoire Painted sites are found in caves or rockshelters rather than in open-air settings, protection from the elements appears to account for some of the difference in frequencies. The occasional presence on

T

142

5 — PURGATOIRE PAINTED ROCK ART

OF THE

HIGH PLAINS

143

canyon walls, however, of remnants of faded figures—too indistinct to identify— suggests that at least some paintings were placed on open-air surfaces but have not survived. Also contributing to differential preservation is the fact that painted figures usually occur on sandstone, whose lighter color was a more suitable background than the basalt on which petroglyphs were more frequently placed. Sandstone surfaces are less resistant to weathering, however, which means that painted sites erode more rapidly than petroglyph sites. The color palette of Purgatoire Painted figures is narrow, consisting exclusively of shades of red ranging from pinkish to brick. Pigment analyses indicate that iron oxides or red ocher provided the color, but the kind of binding agent used in manufacturing the paint is unknown. The majority of the figures appear to have been created by a finger coated with paint, although a few narrow lines may have been made with a frayed stick or a bone stylus. No painting tools of any kind, however, have been recovered from Purgatoire Painted sites. Small, stick-figure humans are characteristic of the Purgatoire Painted style. Seldom more than 15 centimeters tall, these simple figures—red painted in nearly every instance by finger—have little embellishment. Quadrupeds are also common and often resemble Purgatoire Pecked animal figures. Painted and pecked quadrupeds are frequently found superimposed on one another, an association that along with other evidence discussed later in this chapter suggests that the two types are contemporary. A relatively large number of abstract figures occur at Purgatoire Painted sites, although for reasons that are not well understood, some panels have more of these figures than others. Significant numbers of painted abstract figures occur, however, in all painted rock art panels. I begin my detailed discussion of the Purgatoire Painted rock art style with several short descriptions of specific sites and the figures found there. A summary of the characteristics of this important style and a discussion of the contribution made by ethnographic sources to its understanding follow the site descriptions.

THE GAME DRIVE SITE Another of Heywood’s proverbs aptly describes the question rock art researchers face when they attempt to make sense of a landscape containing a number of petroglyph or pictograph panels. What, if anything, goes with what? The answer is self-evident when one looks at a discrete entity such as a painting by Matisse or Manet, but the boundaries defining a unit of meaning in rock art are not always obvious. In fact, the Heywood proverb “you cannot see the wood for the trees” applies to the first efforts by Denver University survey archaeologists to determine whether the rock art at the Game Drive site in Piñon Canyon was unified or a series of unrelated phenomena.

144

L AWRENCE L. LOENDORF — THUNDER

AND

HERDS

In reaching their decision, the team used a common rule of thumb in such situations, based on the distribution of artifacts surrounding a rock art cluster. The convention stipulates that if artifacts—in this case the flaking debris produced by the manufacture of stone tools—are confined within a 20-meter perimeter surrounding a rock art cluster, then those petroglyphs or pictographs are considered a discrete site. By applying this convention, the team decided that they were observing four discrete “trees,” and they divided the area into four individual sites. More recently, Fort Carson archaeologist Mark Owens visited the area and noted that lithic debris was present in an uninterrupted distribution between the areas previously defined as four separate sites. In other words, when Owens looked carefully at the same set of trees, he was able to see that they were actually part of a larger “wood,” to use Heywood’s term. This was an important observation because it motivated researchers to look for possible thematic relationships among the different rock art clusters. From this scrutiny, a model emerged that integrated the area’s different parts into one large game drive complex which had been unrecognized during the more detailed analysis of the individual sites. The Game Drive site is located within a cluster of sandstone outcrops situated along the north bank of Van Bremer Arroyo. The site’s configuration—and its position at the southern edge of the vast grasslands extending north from the Piñon Canyon Maneuver Site into southeastern Colorado—created opportunities for prehistoric hunters to exploit the herds of grazing animals, especially pronghorns, present then and now throughout the year. Closer to the site, the sandstone walls of the eastern and western sides of the drainage converge like a funnel. As the slope of the drainage deepens, the walls constrict, ending in a box-like, 8- to 10-meterhigh outcrop that forms a natural trap. An outlet to the south is almost completely blocked by large boulders resting on the surface of the drainage (Figure 5.1). A single tall rock cairn standing like a sentry on the eastern sandstone wall may be an important prehistoric component of the site. As part of the research at Piñon Canyon, we learned that some original stone monuments were in all likelihood the work of prehistoric Indians and that Euro-Americans may have added stones to preexisting cairns.3 How would hunters have worked together to influence the behavior of their prey during the kind of game drive that gives the site its name? Physical positioning was key, as were the coordinated efforts of numbers of hunters. Some would have been positioned along the tops of the canyon walls. Another person, often a shaman, would have coaxed nearby grazing animals in the direction of the canyon. Pronghorns are extremely curious and are easily attracted to something as simple as a colored cloth waving in the breeze. At just the right moment, the prepositioned hunters would leap up, startling the pronghorn antelope and making them run down the valley, looking for an escape route. In their panic, the antelope would mistake the stone cairn on the ridge to the east for a human, par-

5 — PURGATOIRE PAINTED ROCK ART

OF THE

HIGH PLAINS

145

FIGURE 5.1. The Game Drive site with its constricting sandstone walls and the boulder-strewn outlet to Van Bremer Arroyo. Photograph by Lawrence Loendorf.

ticularly if actual humans were waving their arms and forcing the animals in the direction of the trap. Antelope are not jumpers, so if the hunters had also arranged a row of upright slabs on the ridge—using rocks exposed along the valley’s eastern side—and were running behind them, any attempted escape over the ridge would have been prevented.

Rock Art at the Game Drive Site Fifteen panels of pictographs and petroglyphs occur along the walls of the sandstone outcrop and in two large rockshelters that provide good protection from the elements. The eight rock art panels located in the shelters contain equal numbers of pictographs and petroglyphs. The remaining seven panels occur on the exposed sandstone walls and freestanding boulders in the drainage and contain only petroglyphs, some of which have been enhanced by the addition of red paint. The opening of one of the rockshelters is between 14 and 15 meters wide; it has a maximum depth of 5 meters, and its relatively uniform ceiling height varies between 2 and 2.5 meters. In two dimensions, the second shelter is similar to the first—it has a depth of 5 meters and a maximum ceiling height of 2.3 meters—but its 23-meter opening is considerably wider. The two shelters contain a total of 56 stick-like anthropomorphic figures, 71 quadrupeds, and an undetermined number of abstract figures (Figure 5.2). The pigment used to create approximately three-quarters of the painted figures ranges from a pinkish

146

L AWRENCE L. LOENDORF — THUNDER

AND

HERDS

FIGURE 5.2. A typical panel of pecked and painted figures at the Game Drive site. Note the anthropomorphs chasing quadrupeds and the bird-like figure beneath them. Field drawing by Seija Karki. shade to dark red. In many cases the difference in color appears to be related to the background on which the image has been placed. The anthropomorphic figures are a fairly uniform 10 centimeters tall and consist of a straight vertical line representing the body, crossed by a horizontal line creating two arms. Most often the straight arms project at right angles from the body, but sometimes they point downward. Legs are depicted by inverted Vor U-shapes attached to the lower end of the body line. A few figures have Lshaped feet, but generally neither feet nor hands are indicated. Heads are circular and occasionally rest on a continuation of the body line meant to indicate a neck. More often, however, they are positioned at the intersection of the vertical and horizontal lines representing the body and its arms. A significant number of the figures have a long appendage between their legs which has been interpreted as a penis, but the fact that many of these lines are connected to cracks in the rock surface suggests that they may represent something else (Figure 5.3). Quadruped figures, which can be rectangular, oval, or boat-shaped, normally measure 18 to 20 centimeters from head to tail, although larger examples are sometimes 30 centimeters long. Heads have pointed snouts, and a few figures have branching antlers. Tails are often indicated by straight lines, and the figures lack hooves. Most of these simple, stylized figures probably represent antelope.

5 — PURGATOIRE PAINTED ROCK ART

OF THE

HIGH PLAINS

147

FIGURE 5.3. A painted anthropomorph and quadruped at the Game Drive site. Note that the long appendage between the legs of the anthropomorph connects to a crack in the sandstone wall. Photograph by Lawrence Loendorf.

A single bird-like figure with outstretched wings and downward-pointing feathers has been placed under a ledge below a large panel of quadrupeds and anthropomorphs. There are remarkable similarities between the shape and features of this figure and the design elements typical of other bird images in the region. Variability in this image category is primarily confined to whether the wing feathers point up or down. Abstract image types include short, parallel vertical lines, incomplete circles, curved lines, dots or rows of dots, rakes, and rectangular forms, one of which is a box with an interior X-based design. A large number of the abstract images are, however, paint blobs or smears remaining from once-complete figures. From their position in the panels, it is apparent that these eroded figures represented anthropomorphs or quadrupeds. As for abstract petroglyphs, they occur primarily in exposed positions, where it is possible that they were originally associated with painted representational forms. One panel of abstract figures is located on a sandstone exposure at the apex of the funnel-shaped drainage, near where it empties into Van Bremer Arroyo. Ten abstract petroglyphs occur in an area that is approximately 1.5 meters wide and 75 centimeters high. Five of the figures are meandering forms positioned horizontally across the rock surface and composed of lines that, in some instances, intersect and form enclosed spaces. More often, however, they assume irregular shapes

148

L AWRENCE L. LOENDORF — THUNDER

AND

HERDS

that wander across the surface. The individual petroglyphs are between 20 and 40 centimeters long, but they may have once been part of a much larger meandering form. The panel’s remaining elements include a tailed circle, a dot, a single line, a partial circle, and several areas of apparently random pecking. Three additional petroglyph panels—which include rectangular box forms, grid patterns, and connected circles—occur on the boulders that obstruct the exit from the drainage into Van Bremer Arroyo. Another nearby panel contains an intriguing set of 11 vertical lines that are crossed near their upper ends by a horizontal line or bar. Six of these vertical lines are also connected to a semicircle or loop at their lower ends. One of the vertical lines intersects an ambiguous figure that may represent a quadruped and is the only arguably representative image in the panel. In Chapter 4, I discussed the propensity of rock art researchers to identify any linear, handheld rock art implements as atlatls. This interpretation has been suggested for the 10- to 20-centimeter vertical lines at the Game Drive site that end in bisected circles. The circular loops are said to represent atlatl handles, and the crossbars are claimed to replicate atlatl weights. Fortunately, there are some useful resources for evaluating this proposition. In rockshelters along the Pecos River in southern Texas, large multicolored pictograph figures hold equally convincing atlatls which are frequently attached to a dart or spear, as if launch was about to occur.4 These atlatls are defined by curved lines to which no weights or hooks have been added. Petroglyphs thought to represent atlatls are relatively common at the Jeffers site in southern Minnesota.5 Their shapes vary, but many consist of a vertical line, presumably the atlatl shaft, which passes through an open circle. The shaft lines of other figures pass though a rectangular or circular solidly pecked area that resembles the atlatl shields used by Australian aborigines. Many of the Jeffers figures also have hooked ends. While I am not rejecting out of hand the possibility that the linear-looped images at the Game Drive site are atlatls, an alternative interpretation is that they represent the thrusting spears known to have been used by hunters to dispatch animals caught in nets. Resolution of the ambiguity surrounding these linear images will depend, however, on expanding the number of known instances of such figures in the rock art record, coupled with careful attention to the contexts in which they occur and the other figures found with them.

The Game Drive Site Rock Art as a Hunting Scenario By integrating what is known about the physical setting at the Game Drive site and the placement and content of the site’s rock art, I believe a strong argument can be made for a scenario in which prehistoric hunters were successfully driving grazing animals into a trap and—either before or afterward—placing engraved and painted hunting-related images at the scene (Figure 5.4). There is precedent

5 — PURGATOIRE PAINTED ROCK ART

OF THE

HIGH PLAINS

149

FIGURE 5.4. A contour map of the Game Drive site. Illustration by Thaddeus Swan. for such an interpretation. Landform features at other sites in the region are associated with deliberately placed rock art panels containing quadrupeds and netlike motifs depicting hunt-related activities. All hunting-related sites are not

150

L AWRENCE L. LOENDORF — THUNDER

AND

HERDS

alike, however. At most sites in the region, the juxtaposition of animal figures and nets is concentrated in one panel, with quadrupeds shown on the opposite side of a net from human figures. At the Game Drive site, however, the animals are more scattered, but this difference is mitigated by the fact that the net-like elements occur at precisely the location where an actual net would have been placed to capture animals (Figure 5.5). Also important is the similarity between the grid forms at the Game Drive site and those at Purgatoire Pecked-I sites, which are also thought to represent nets or snares. Why petroglyphs depicting nets and snares were placed at the very point in the drive complex where their real counterparts would have been most effective is an intriguing question. One possibility is that these motifs had a magical function and that their power was intended to impede or stop the game from escaping. Alternatively, the rock art panels may have had a pedagogical purpose, indicating to hunters-in-training where the nets essential to a successful game drive were to be stretched. At other sites in the region, images of hunters or drivers actively engaged in herding animals provide sufficient indication that some rock art is hunting-related. At the Game Drive site, however, there are no panels where humans actually chase animals into nets or snares. It would appear that it was more important to put all the essential elements of a hunt—quadrupeds, humans, and hunting equipment represented by looped-line motifs—and the position of grid forms near or adjacent to animals and humans in the panel than it was to create a replica of hunting activity.

FIGURE 5.5. The loop-line and net-like figures at the outlet of the Game Drive site. Drawing by Seija Karki.

5 — PURGATOIRE PAINTED ROCK ART

OF THE

HIGH PLAINS

151

The historical literature of communal hunting frequently includes descriptions of individuals who control and coordinate the hunt. The inclusion of a prominently featured bird-like figure may be an allusion to a person who goes into trance and locates game during an out-of-body experience. Owls and other nocturnal species, as well as eagles and hawks across the Great Plains, are often considered to be the tutelary spirits of hunt masters because they can fly over large territories and see for great distances. It is in the rockshelters of the Game Drive site that most of the quadrupeds and anthropomorphs co-occur in scenes with abstract figures, some of which may represent nets or snares. A tally of the figures in the eight panels in these shelters provides the following totals by category: 7.8 percent are circles or circular and curvilinear forms; 16.5 percent are dots and dot forms; 21 percent are lines, grids, and rectilinear forms; 30 percent are quadrupeds; and 24 percent are anthropomorphs. When these percentages are compared with similar tabulations of figure types for various time periods in the region, their place in the seriation (Figure 1.7) indicates that the Game Drive site is slightly older than the Corral site and somewhat more recent than the Zookeeper site. In calendar years this is about A.D. 1300, but it must be pointed out that this estimate is based only on the two main areas of rock art at the site. Those panels containing a small number of often-eroded abstract figures may represent an earlier use of the site.

THE ROCK CROSSING SITE Looked at from the air, the portion of the High Plains in which the Piñon Canyon Maneuver Site is situated is anything but plain. The surface appears creased and rumpled, scarred by a network of twisting, branching channels that begin as small, ephemeral drainages and grow to large arroyos feeding the Purgatoire River. The view from above suggests a landscape that is best negotiated on foot or by horseback. Motor vehicles are at a disadvantage in such terrain, and the distribution of human dwellings reflects this fact. The ranches and old homesteads that are situated in canyon country were established during the days of horse and buggy, while ranch houses located on the grasslands, away from arroyos, postdate the invention and widespread use of automobiles. For many years, the problem was how to get from here to there. Arroyo crossings often have muddy bottoms that can immobilize the wheels of cars and pickup trucks. In places like the Piñon Canyon area, roads invariably follow a route requiring the fewest crossings. These crossings occur at those locations where the upland terrain slopes gradually into a bedrock bottom, forming a passage often referred to as a rock crossing. One such rock crossing is located just below the junction of the Burke and Taylor arroyos on the western edge of the Piñon Canyon Maneuver Site. From the perspective of a hawk looking for its dinner upstream from this junction, the

152

L AWRENCE L. LOENDORF — THUNDER

AND

HERDS

many small drainages feeding into each arroyo look like capillaries merging to form a vein. Downstream, the combined stream course cuts into the bedrock sandstone to create a formidable canyon. To someone behind the wheel of a 4×4, conditions at both ends of the channel system would make crossing impossible, which is why in an aerial view of the rock crossing, roads heading through it to the north, south, east, and west are clearly visible. Given the location’s strategic position in the surrounding countryside, it is predictable that it would contain a major archaeological site known, not surprisingly, as the Rock Crossing site. Extensive evidence of prehistoric and historic use is concentrated around nine rockshelters or sandstone overhangs that occur along the bedrock bottom to the west of the main crossing. Several of the rockshelters have attached sandstone slab walls that form corrals or holding pens once used for sheep and goats. Test excavations near and in the rockshelters have revealed that one shelter—containing a stone fireplace—was used as a habitation during the late 1800s and early 1900s. The Spanish surnames painted on one wall, probably with sheep-branding paint, suggest that at least some of the site’s inhabitants were Mexican sheepherders. Excavators have also recovered considerable prehistoric debris at the site, including manos and metates intermixed with heat-cracked stones and charred soils.6 Dozens of chipped stone flakes and a variety of penetrating, cutting, and scraping tools were also found. Burned fragments of small mammal bones identified as the remains of rabbits and prairie dogs occurred in the test pits, but the artiodactyl species from which larger bone fragments came could not be determined. No ceramics were found at the site, and there are no radiocarbon dates. Some idea of the time periods during which the site was used comes from the presence of Scallorn corner-notched projectile points and other arrow points that are securely dated to the Developmental and Diversification Periods.

Rock Art at the Rock Crossing Site The prehistoric rock art at the site consists of 37 pictographs and three petroglyphs. The site’s pictographs, measured across their maximum width, range from 8 to 16 centimeters and are consistent in size and simplicity of design with others assigned to the Purgatoire Painted style. Pigment analysis indicates that the red to reddish-orange paint used to define the pictographs came from red ocher containing large amounts of hematite.7 The associated petroglyphs are solidly pecked and resemble those found at nearby Purgatoire Pecked sites. Most of the pictograph and petroglyph figures occur in a single rockshelter that contains three panels—one on the back wall, one on the ceiling, and one prominently placed on the outer face of a narrow shelf of sandstone exposed along the ceiling near the shelter’s opening (Figure 5.6). The figures on the back wall include a pecked quadruped, two painted quadrupeds, three painted stick-like

5 — PURGATOIRE PAINTED ROCK ART

OF THE

HIGH PLAINS

153

anthropomorphs, and a painted figure that resembles a turkey. The quadrupeds, none of which have antlers, are simple figures with relatively stubby legs and pointed snouts. The anthropomorphs are also simple figures, shown in full-frontal view, with upright bodies and bow legs. The position of the arms varies: they are pointed up on one figure, down on another, and are indicated by a straight, horizontal line that crosses the body of the third. None of the arms have digits, and none of the figures have headgear. The turkey-like figure has a downward-pointing head, two legs, and an upraised tail (Figure 5.7). It is not as well-made as a similar figure at Renaud’s Shelter (discussed below) near Colorado Springs, but its resemblance to the latter figure is sufficiently strong for me to feel confident that the image can be reliably identified as a turkey. Twenty-nine short, parallel counter lines occur on the ceiling panel of the rockshelter, and there is also a painted anthropomorph whose body consists of a single straight line to which inverted, V-shaped legs are attached. The figure’s arms, which have no attached digits, are indicated by a horizontal line crossing its upright body. Another figure on the ceiling, formed by two triangles that are

FIGURE 5.6. Abstract pictographs on the narrow ledge above the opening of the Rock Crossing rockshelter. Re-drafted by Elaine Nimmo from an illustration by Anne Hayes.

FIGURE 5.7. The pecked quadruped and a turkey-like bird at the Rock Crossing site. Re-drafted by Elaine Nimmo from an illustration by Janet Lever.

154

L AWRENCE L. LOENDORF — THUNDER

AND

HERDS

connected at their apex, is shaped like a bow tie. The ceiling of the rockshelter has not been exposed to the elements, and therefore the sediments are lighter in color than nearby sandstone surfaces. The lighter background has the effect of making the paintings appear more orange than others at the site.

THE BEAR DANCE SITE Archaeologists find sites by surveying. Survey strategies may vary, but the one at Piñon Canyon involves persons walking in a straight line across a predetermined survey unit, eyes cast downward, looking for archaeological remains. A crew chief usually lines up members of the team at intervals along one boundary of a defined area. They set off and, when they reach the opposite boundary, they reverse direction, move a specified distance to the left or right, and then make another sweep across the area, returning to where they started. This approach enables archaeologists to examine a land surface in considerable detail, which makes it puzzling that sometimes the most obvious archaeological evidence is overlooked. Mistakes do not happen often, but some occur because a team does not begin their survey exactly where they ended their search the previous day. Sometimes oversights occur because a team is so focused on the ground surrounding their feet that they fail to notice sites above the plane of their scrutiny. This happens to rock art sites more often than it should. For whatever reason, a major site was overlooked during the original survey work at Piñon Canyon in 1988. Not until 2005, when archaeologists returned to examine a nearby site, did Kendra Rodgers, a sharp-eyed, energetic archaeologist, spot a cave opening and go to take a closer look. The site Kendra found is a low cave that opens to the east, just below the caprock. Its interior measures 9.2 meters across the opening and is 4 meters deep (Figure 5.8). A partially intact rock wall on the southern end of the rockshelter is all that remains of an enclosed structure. Like other walls commonly found at caves and rockshelters in the region, it is made of slabs of sandstone that are set upright to form a low barrier across the cave opening. Although no direct evidence remains, it is possible that this and other rock walls supported a wooden framework that totally enclosed a portion of the rockshelter. Six chipped stone flakes, one made of hornfels or silicified basalt and five made of argillite, were noted on the cave floor (Figure 5.9). Three argillite stone tools were also present: an exhausted core, a retouched flake, and a scraper. One slab metate fragment was found under the rubble from a collapsed portion of the northern end of the wall. A utilized argillite flake associated with the wall was near the slabs that remained vertical. A bedrock metate (Feature 5) was situated on a low shelf at the southern end of the main rockshelter. Of its four milling surfaces—three of which are oval and one of which is irregularly shaped—only three show signs of moderate grinding

5 — PURGATOIRE PAINTED ROCK ART

OF THE

HIGH PLAINS

155

FIGURE 5.8. A view of the Bear Dance site showing the rock wall that encloses the rockshelter. Note the large upright sandstone slabs in the wall. Photograph by Kendra Rodgers.

FIGURE 5.9. A plan-view map of the Bear Dance rockshelter site. Illustration by Kendra Rodgers.

156

L AWRENCE L. LOENDORF — THUNDER

AND

HERDS

(one with pecking), and none have striations. Marks made by heavy grinding and pecking are visible on the surface of another metate—which is 23 centimeters long, 18.5 centimeters wide, and 2.2 centimeters deep—and it is the only grinding surface with a notable depth. A third bedrock metate (Feature 6), positioned on a sandstone ledge along the back wall of the rockshelter, is characterized by light grinding in an irregularly shaped area that measures 24 by 17 centimeters. Despite the proximity of these metates to the pictographs, there was no visible evidence that they had been used for processing pigments.

Rock Art at the Bear Dance Site As Kendra Rodgers approached the Bear Dance site for the first time, she noted some eroded paintings on the canyon wall outside the cave, but it wasn’t until she went inside and looked up at the ceiling that the significance of her discovery became apparent. There she found several scenes consisting of red pictographs, each containing small, stick-like human figures interacting with the figures of bears (Figure 5.10). In one scene, a bear is depicted with a spear in its back, surrounded by small, spear-carrying human figures. Unfortunately, roof fall has taken its toll on the painted ceiling in a large area between the two panels, and paintings may have been destroyed. In another scene, to the right of the damaged area, the theme of stick-like human figures with spears attacking a bear is repeated. This panel also contains a large plant-like form and a crescent-shaped figure with protruding lines, like eyelashes, along its upper edge. An especially interesting human figure is prominently positioned immediately in front of the bear and appears to be wearing a headdress that replicates the design of the plant-like form (Figure 5.11). Another

FIGURE 5.10. A scale drawing of the Bear Dance Panel. Illustration by Kendra Rodgers.

5 — PURGATOIRE PAINTED ROCK ART

OF THE

HIGH PLAINS

157

FIGURE 5.11. The dance scene in the Bear Dance panel. Photograph by Lawrence Loendorf. eroded area to the right of this scene contains the back and ears of a figure that appears to be another bear, also with a spear stuck in its back. This bear image may be associated with a rayed arc that occurs above it or, alternatively, any figures originally relating to it may have eroded from the ceiling panel.

Placing the Bear Dance Site in Time No chronometric dating has been attempted on any of the figures at the Bear Dance site or on any other images included in the Purgatoire Painted style, so relative dating techniques must be used to establish the age of these paintings. Normally, the presence of diagnostic artifacts would help to determine when activities took place at the site, but, unfortunately, no objects were found on the surface of the cave floor. A general idea of the chronological placement of the site can be deduced, however, from the fact that the partly intact rock wall made of horizontally stacked sandstone blocks includes several prominent slabs set upright to form a protective barrier. This kind of vertical orientation is common in archaeological sites associated with the Apishapa culture, which is dated between A.D. 1050 and 1450 in southeast Colorado. Several types of figures at the Bear Dance site have attributes linking them to a specific time period. The small human stick figures with V-shaped crotches and L-shaped feet are quite similar to anthropomorphs associated with game drive scenes at Diversification Period rock art sites. The depiction of human figures in profile view is another attribute characteristic of the Diversification Period, and the images of bears with straight legs and downward-pointing claws are similar to petroglyphs of bears from sites during the same time period (compare with Figure 4.16). The consistency demonstrated by these similarities is evidence that the Bear Dance site should be provisionally dated to the Diversification Period.

158

L AWRENCE L. LOENDORF — THUNDER

AND

HERDS

Who Was Dancing at the Bear Dance Site? The alert reader is undoubtedly wondering how the Bear Dance site got its name. From the descriptions of the site’s rock art scenes that I have provided, it would be logical to conclude that they represent straightforward depictions of a bear hunt. After a careful consideration of relevant ethnographic sources, however, I think it is more likely that another set of dynamics is being expressed in the panels. My primary source is Dr. Gene Weltfish, an anthropologist whose book, The Lost Universe, chronicles her fieldwork in the 1930s with Caddoanspeaking Pawnee tribes who were descendants of tribal peoples of the eastern Great Plains. Weltfish’s linguistic and cultural studies gave her access to many aspects of contemporary Pawnee life, including important events in the group’s ceremonies. She was particularly interested in the organization and role of spiritual and physical healers, who were referred to as “doctors,” and whose “Doctor Association” coordinated aspects of the annual ritual cycle. The ceremonies were more than religious observances. They were the whole focus of Pawnee aesthetic life, particularly in the performing arts. The pageantry and the costuming, the dances and miming were developed for beauty as well as for religious significance. They were opera and ballet, and the songs were appreciated for their technical and aesthetic value.8 Weltfish’s appreciation for her subject helps one overlook the anachronistic term—the “Thirty-day Grand Opera of the Wild Animal Cults”—that she uses to describe one particularly important event. Animal cults, which had their origin in the visions of individual men, played important roles in the “Grand Opera” of the Pawnee doctors that Weltfish described, and its much anticipated grand finale included a contest between bears and buffalo clown warriors. As she reported, a Pawnee doctor had received special knowledge that would help him cure the sick through the properties of the plants and other curative procedures. In addition to this kind of knowledge, the animal also transmitted to his protégé the power of hypnotism and suggestion that gave him control over animals and other people and, if he was skillful enough, over enemies of the tribe.9 The ritual knowledge required to become a doctor in an animal cult was handed down from a practicing doctor to an apprentice, generation after generation, with some modification through time. Members of animal cults officiated in individual ceremonies at different times throughout the year, but they all came together at a large “Doctor Lodge” to participate in the “grand opera.” The theatrical event took place in a sponsor’s earth lodge, from which all furniture and personal belongs had been removed. Once the lodge had been swept clean, members of the animal cults constructed booths along the sides of the building.

5 — PURGATOIRE PAINTED ROCK ART

OF THE

HIGH PLAINS

159

The wood and lattice booths had only one small opening which was used as an entrance and exit by the persons belonging to the cult. Weltfish refers to the Doctor Association as “long established,” having been founded by a tribal visionary who is honored by the construction, inside the earth lodge, of large images commemorating his vision.10 The largest of these was a three-dimensional 60-foot-long water serpent that wound around the lodge floor. Its frame was made of ash wood and covered with grass, over which a layer of clay was plastered and then painted in various colors. The serpent’s mouth was large enough for a human to crawl inside. A second figure represented the witch woman who, according to legend, lived in the animal lodge under the Missouri River where the founding visionary was taken in his trance. This realistic figure had long braids and wore a buffalo hide dress. Additional figures included a fireplace in the form of a turtle which was molded from clay and occupied the center of the lodge. Cutouts of a star god and various animals were made of buffalo hide and suspended from the ceiling of the lodge. The preparatory work took several days, after which the participants went in search of a sacred cedar tree that was placed just northeast of the door to the bear cult booth. This practice represented the linkage of bears and cedar trees and the fact that their ritual direction was northeast. Bears and cedar trees also had celestial counterparts in the sun and the moon. The ceremonies sponsored by the Doctor Association took place in a prescribed sequence. A number of outdoor dances were followed by theatrical performances inside the lodge. Using magical charms and imitating the movements of the animals they represented, members of various animal cults practiced putting rivals into hypnotic trances. They also displayed such convincing sleight-ofhand tricks that the white traders and army personnel who attended these ceremonies were completely baffled about how the magic worked. The events of the final ceremony, in which the bear shamans personified their cult animals by dressing in bearskin costumes and wearing cedar wreaths around their necks, are the most relevant to the rock art panels at the Bear Dance site. Interacting energetically, clown dancers and shamans participated in a drama meant to teach a lesson to the cult members portraying young bears in the booth, who were being guarded by older bears. Wearing headdresses of corn husks, the spear-carrying clown dancers represented warriors, and their leader had a shield, also made of corn husks. Prodding vigorously with their spears, the clown dancers would force the young bear impersonators out of their lodge, after which a great deal of dramatic chasing and jumping took place. Eventually the bear leader would attack the younger bears and anything else in his path, such as house posts, by swinging a stone attached to a rope. After the younger bears had been knocked down and subdued, the drama moved outside, leaving the doctors in the lodge, where they

160

L AWRENCE L. LOENDORF — THUNDER

AND

HERDS

used a lighted pipe to revive the bear impersonators and return them to their booth. The ceremony ended with the emergence from the booth of all of the bear doctors, no longer wearing their costumes. Although several components of the Pawnee bear ceremony correspond to aspects of the paintings at the Bear Dance site, we must remember that Weltfish’s account is idealized and represents the way the ceremony might have taken place before significant cultural alterations occurred. Nevertheless, I believe that the figures at the Bear Dance site almost certainly were placed there in the context of ancestral Pawnee bear power. Support for this claim is based on the association of several important figures in the panel. The first, on the right side, is the painted, crescent-shaped figure resembling a rising sun, from which stick-like lines protrude. Early twentieth-century Anglo-Pawnee scholar James R. Murie, who worked under the supervision of the noted anthropologist Clark Wissler, has explained that in several Pawnee accounts, bears are associated with the morning sun. “The sun would appear in the east, where all the bear family had to look to; for from the sun they received their power of animal magnetism and also inhaled the rays of the sun so that no harm would come to them”11 Murie also noted that the original man or hero of the bear society was told that “whenever he wanted to replenish the dust he must get it from the sun at dawn by inhaling its rays.”12 It is likely that the panel’s orientation in an east-facing shelter that receives morning light was purposeful and is related to the bear’s receiving power from the rising sun. The large plant-like figure in front of the rock art image of the sun seems to correspond to the sacred cedar tree that was prominently placed in the Doctor Lodge in the Pawnee ceremony. Support for the association of bears and cedar trees is found in Murie’s account of the origin of the bear dance, which describes how, after many days of fasting, Roaming Chief—the hereditary chief of the Chaui Pawnee—had a visionary experience. In a wooded, hilly area he heard a woman singing and discovered that the sound came from a young woman who stood next to a cedar tree, facing the sun. He noticed that “she had upon her head a young cedar sprout and carried in her left hand a small limb of a cedar tree.”13 The young woman taught Roaming Chief about the powerful relationship between the cedar tree and the bear, and later, whenever he displayed his power, he wore a cedar branch on his head. Another significant correspondence between the Pawnee ceremony and rock art at the Bear Dance site relates to the figure of the bear placed in front of the image of the sun. Its downward-pointing, oversized claws and its flat back suggest the image is meant to represent a human wearing a bearskin costume. As part of the pageantry of the ceremony in the Doctor Lodge, a bearskin covered a shaman’s body and a cedar wreath encircled his neck. The wreath not only reinforced the ritual association between bears and cedar trees but, more practically, it helped to keep

5 — PURGATOIRE PAINTED ROCK ART

OF THE

HIGH PLAINS

161

the robe in place on the shaman’s back. In the rock art panel, a small darkened band around the neck of the bear figure may represent the cedar wreath. The human figures in the rock art panel that appear to be carrying spears may be playing a role similar to the one performed by buffalo warriors in the Pawnee ceremony. It is particularly significant that the human figure standing in front of the bear image is wearing a headdress that replicates the pattern of the branches of the cedar tree image. In Weltfish’s account of the animal cult ceremony in the Doctor Lodge, clown warriors wore cornstalk headdresses but, as described by Murie, cedar branch headdresses were the customary headgear of bear shamans. Changes in ceremonialism during the 500 years separating the Pawnee event described by Weltfish and the creation of the paintings in the Bear Dance panel could, however, account for the change from cedar branches to cornstalks. The other portion of the panel, which contains small, spear-wielding human figures that appear to be attacking a bear, may also be related to Pawnee bear shamanism. The separation between the scenes, one of which is more elaborate than the other, may indicate that two separate parts of the animal cult ceremony are being depicted. The second panel, for instance, might represent cult members who are chasing a bear, either outside of its booth or outside the Doctor Lodge altogether. On the other hand, since the ingredients of the paint in the two scenes are different, one of two conclusions is likely: that the panels were painted on different occasions, or that the second panel repeats the events depicted in the first, but with fewer details. Regardless of the precise relationship between the scenes in the two panels, the paintings at the Bear Dance site are almost certainly related to the Pawnee ritual associated with bear shamanism. Several circumstances may account for the presence of the paintings on the ceiling at the Bear Dance site. They may have been recorded by a cult member after a vision in which a bear was killed by clown warriors. In Weltfish’s account, the large three-dimensional figures in the Doctor Lodge were replicas of phenomena originally observed in a vision by the founder of an animal cult. These more recent Pawnee creations, made of willow frames covered with clay, may be intended to remind cult participants of the content of the vision and therefore to be part of a tradition that has great time-depth. It is also possible that only the very detailed composition of the right-hand panel may reflect this kind of visionary experience, and that another set of actions is depicted in the panel on the left. It is also possible that the rockshelter might have been a place where bear shamans performed healing ceremonies during which paintings were created as part of the curative process. The presence of well-used bedrock metates may be an indication that medicinal plants were prepared there. Alternatively, the paintings may have been produced as part of a ceremony in which bear power was transferred to an apprentice. It is well documented that Pawnee shamans taught their

162

L AWRENCE L. LOENDORF — THUNDER

AND

HERDS

power to younger individuals, and the teaching experience may have involved the depiction of images related to the original vision. One thing is certain, however. Excavation of the rockshelter’s deposits would help determine the kind of the activities that took place in the shelter in the past and might also indicate whether there is a connection to the ceremonial life of more contemporary peoples.

“A HARD BEGINNING MAKETH A GOOD E NDING” There was much to learn when I began my research into the attributes, distribution, and chronology of the Purgatoire Painted rock art style. Heywood’s “hard beginning” was an appropriate description of the scale of the work required by this fascinating topic. Thankfully, in the past 20 years, my team and I have learned enough to be able to make some definitive statements, and some intriguing provisional ones, about what these painted figures represent. For example, it is possible to say with confidence—based on the overlap between Purgatoire Painted and Purgatoire Pecked-II sites, and the general proximity of Purgatoire Painted sites to the distribution of Apishapa village sites—that these painted figures are associated with the Apishapa peoples. And even though no Purgatoire Painted pictograph has been directly dated, using the relative chronological controls in place for Purgatoire Pecked-II petroglyphs, I have considerable confidence that the paintings originated in the Diversification Period, or between A.D. 1050 and 1450. I have consulted the ethnographic record on numerous occasions in my effort to understand some of the cultural practices that might account for the apparent narrative character of the pictographs at the Bear Dance site. The commonalities between those painted panels and Pawnee bear dance ceremonialism are so striking that I believe there is almost certainly a relationship between them. One very important outcome of this linkage is that it gives me confidence in the use of Caddoan ethnography as a body of information for the study of central High Plains rock art. As I note elsewhere, archaeologists had already concluded that Caddoan groups were descendants of Apishapa peoples, but the evidence at the Bear Dance site provides considerable support for that inference. The presence of painted rock art images of turkeys is another link to Caddoan traditions. Based on an extensive study of Pawnee ethnographic sources, anthropologist Patricia O’Brien has noted the significance of turkeys as the original protectors of children.14 She reports that in the dualistic Pawnee world view, the Morning Star—the source of male power—was symbolized by eagles, and the Evening Star—associated with conception—was originally represented by the turkey and later by the bobwhite quail. It is also relevant that the daughter of Evening Star used turkey feathers to help her escape from the Pawnee ogre, Long Tooth. O’Brien points out that the turkey and bobwhite quail are logical choices to represent women because they are ground birds and in constant contact with the earth, which in Pawnee tradition is conceptualized as female.

5 — PURGATOIRE PAINTED ROCK ART

OF THE

HIGH PLAINS

163

The co-occurrence at Purgatoire Painted sites of images of turkeys and abstract forms is also significant. In a seminal study of Plains Indian art, ethnologist John C. Ewers recognized that geometric designs were consistently produced by women and life forms were painted by men.15 This pattern extended to the use of chevrons, triangles, and other abstract patterns on Plains ceramics, traditionally the domain of female potters, and it may account for the types of figures occurring at rock art sites on the central High Plains. The Rock Crossing site contains primarily abstract forms, along with lesser numbers of anthropomorphs and quadrupeds. Designs at Renaud’s Shelter, between Colorado Springs and Canon City, Colorado, include a painted turkey, two well-made female deer, and an abstract pattern that resembles the figure often painted on parfleches, or rawhide bags (Figure 5.12). If women processed deer hides at Renaud’s Shelter, then their presence could account for the use of all three symbols. The placement at the Red-tail site of abstract painted figures in the kitchen area—traditionally women’s work space—and their separation from scenes of communal hunting add further support for the linkage of image type with gender. Purgatoire Painted figures are found in a relatively restricted geographical area. They occur most frequently in the Purgatoire River canyon itself and in numerous small rockshelters in side canyons entering from the west. Paintings have also survived in a group of rockshelters in the Colorado piedmont, near the mountains south of Colorado Springs, where they typically occur on the outer surfaces of narrow ledges near the top of low sandstone overhangs. At other sites, such as McNees Crossing, north of Clayton, New Mexico, Purgatoire Pecked-II petroglyphs have sometimes been filled with red paint. Because these sites also

FIGURE 5.12. The deer, turkey and parfleche-like design from the Renaud rockshelter, Turkey Creek, Fort Carson, Colorado. Re-drafted by Elaine Nimmo from a field drawing by Margaret Van Ness.

164

L AWRENCE L. LOENDORF — THUNDER

AND

HERDS

contain painted figures that are not part of the Purgatoire Painted tradition, it is not clear whether the painted petroglyphs were the work of Apishapa peoples or of some other group. When Cole originally defined the characteristics of the Purgatoire Painted style, the boundaries were more extensive and embraced a wide range of painted figures, among them images of the large black bears found throughout the region. During the time when the definition of the Purgatoire Painted style was still a work in progress, I also included several types of painted figures that I no longer think belong in the category. Confusion particularly surrounded the appropriate typological placement of some narrow-waisted human figures that appear to be wearing garments. Cole noted the similarities between these figures and Apache painted figures occurring in southern New Mexico and then concluded that Purgatoire Painted and Purgatoire Pecked-II styles were part of the Apache rock art tradition.16 More recent studies have shown that more than one group of painted figures occur on the central High Plains. It is likely that the narrow-waisted figures belong to the Apache stylistic tradition and should no longer be considered part of the Purgatoire Painted style. I briefly discuss Apache rock art in the final chapter of this book. It has been suggested that once the parameters of the Purgatoire Painted style are understood, the painted figures might best be combined with the Purgatoire petroglyphs into a single Purgatoire style. This makes considerable sense, and may well happen with future research.

NOTES 1 All of the proverbs quoted in this chapter have been taken from Heywood (1906, Parts I and II). 2 Cole 1985:17. 3 Loendorf 1989:271–75. 4 Boyd 2003:33. 5 Wellmann 1979:150. 6 Andrefsky et al. 1990:155–82. 7 Loendorf 1989:318–32; Dean 1991:1. 8 Weltfish 1965:8. 9 Weltfish 1965:272. 10 Weltfish 1965:39. 11 Murie 1984:323. 12 Murie 1984:334. 13 Irwin1996:148. 14 O’Brien 1991:59–60; O’Brien and Post 1988. 15 Ewers 1939:7. In a study of shoulder patterns on Plains Indian pottery, W. Raymond Wood

(1962) reported the same gender-based dichotomy in ceramic designs.

16 Cole 1985:71.

6 ROCK ART OF THE PROTOHISTORIC PERIOD A.D. 1 4 5 0 – 1 7 2 5

rotohistoric is a word that we all intuitively understand. I usually think of it as meaning “just prior to written history.” Since, however, I am using the word to designate a 250-year time period in the flow of events on the High Plains, I thought I would consult a dictionary to see if my grasp of its meaning needed refinement. That arbiter of usage, The Oxford English Dictionary, lists the Greek root proto- followed by numerous suffixes including historic, which was defined as “belonging or relating to primitive history, or the beginnings of historical record.”1 Not only did the word primitive make me uneasy, I also felt that the definition was somehow incomplete. It was when I discovered the following statement in the online encyclopedia Wikipedia—“protohistory refers to a period between prehistory and history, during which a culture or civilization has not yet developed writing, but other cultures have already noted its existence in their own writings”—that I knew I had found the nuance I was searching for.2 This chapter examines the arrival and dispersal on the High Plains of the Athapaskan-speaking groups now identified as Navajo and Apache. Beginning in approximately A.D. 1500, a time when the lifeways of Native people did not

P

165

166

L AWRENCE L. LOENDORF — THUNDER

AND

HERDS

include a need for literacy, small groups of Athapaskan-speakers are thought to have begun occupying the territory recently vacated by the Apishapa peoples whose rock art I have described in preceding chapters. By the 1600s, however, Spanish explorers, missionaries, and colonists were arriving in the region and observing its indigenous people. While it is true that they viewed Native peoples through the prism of European cultural traditions, it is in their written records, rather than in the archaeological record, that the most convincing evidence of the protohistoric presence and activities of Apache on the central High Plains is found. Researchers generally agree that the ancestors of the Apache migrated southward from the Mackenzie Basin in northwestern Canada into Montana and Wyoming between A.D. 800 and 1000.3 The cause of the relocation of such large numbers of hunter-gatherers is thought to have been the widespread environmental degradation that followed a massive volcanic eruption in approximately A.D. 800. Several hundred thousand square kilometers in southeastern Alaska and northwestern Canada were inundated with volcanic ash—referred to as White River tephra—which had a severe impact on the availability of human food resources. It is thought that the ash was released by an explosion of Mount Churchill, a stratovolcano in the Wrangell Mountains of Alaska and the Yukon Territory. While there is unanimity among archaeologists about the Athapaskan place of origin and the date when their southward migration began, the routes and the timing of their movements into the American Southwest are robustly debated. One migration model proposes a route through the interior valleys and foothills of the Rocky Mountains, while another suggests that some Apache groups followed a route taking them out onto the Colorado High Plains.4 Alternatively, Athapaskan cosmology places the emergence of its people in a region along the San Juan River in northwestern New Mexico called the Dinetah.5 In an excellent summary of the two divergent models, Ronald Towner identifies one as the “Early Entry, Mountain Route Refugee Hypothesis Model” and the other as the “Late Entry, High Plains Refugee Hypothesis Model.”6 According to the Mountain Route model, once Athapaskan groups crossed the then-nonexistent boundary between Canada and the United States, they traveled southeast through Idaho, Montana, Wyoming, and Colorado to the Dinetah of northwestern New Mexico, where they arrived at least by A.D. 1500, although possibly earlier. A much different scenario is advanced by advocates of the Late Entry, High Plains route. They believe that in roughly A.D. 1450, Apache groups emerged from the Rockies, traversed Wyoming’s Black Hills, and proceeded east onto the High Plains. During this highly mobile phase, subsistence was based primarily on buffalo hunting, which provided a commodity they traded with the horticultural Caddoan-speaking tribes living along the Dismal River in west-central Nebraska. By A.D. 1675, the Apache are believed to have become

6 — ROCK ART

OF THE

PROTOHISTORIC PERIOD

167

semi-sedentary horticulturists who made pottery and lived among the Caddoans until the Comanche disrupted their way of life in approximately A.D. 1725. James Gunnerson has termed the Apache horticultural period the Dismal River Aspect, which he maintains can be identified in western Nebraska, western Kansas, and across eastern Colorado into the Arkansas River basin. To support claims of the presence of Plains Apaches in this region, he and other researchers rely on Spanish documents and diaries that record seventeenth- and early eighteenth-century expeditions into a region that the Spanish referred to as El Cuartelejo, or the far quarter. The accounts describe Apache peoples in El Cuartelejo—which extended from Horse Creek in Crowley County, Colorado, to Scott County, Kansas—living in “rancherias” or small communities of buffalo hide tipis.7 Many rancherias were situated near fortified mesas which in some cases appear to have been former Apishapa strongholds. Spanish documents describe the Apache inhabitants of the region as bison hunters who used dogs to pull travois when they moved from one rancheria to another. Weaponry is also described, including large, sinew-backed “Turkish” bows, flint-tipped arrows, and leather shields so large that they covered a man’s body. The information about protohistoric Apache groups provided by seventeenth-century Spanish documents is supplemented by images in two polychrome paintings on hide created during roughly the same time period by an unknown artist. In 1761, the paintings—known as Segesser I and II—were sent from Mexico by the Jesuit missionary Philip Segesser von Brunegg to his family in Switzerland,8 where they remained until acquired by the State of New Mexico in 1988. The Segesser II painting depicts a battle that took place on August 13, 1720, at the confluence of the Loup and Platte Rivers in present-day Nebraska. During the conflict, a group of Pawnee and Oto Indians, assisted by French forces, attacked a Spanish expedition led by Pedro de Villasur, killing 33 Spaniards and 12 of their Pueblo allies. More relevant to my interest in Apache protohistory is the Segesser I painting, created between 1720 and 1729. Segesser I is apparently a representation of one of the many punitive expeditions undertaken by the Spanish in retaliation for Apache raids on settlements in New Mexico (Figure 6.1). It shows a group of mounted Pueblo genízaros (the offspring of Spanish colonists and indigenous Pueblo people) armed with Spanish weapons and attacking an Apache tipi village, which is situated next to a defensive rock breastwork protecting the women and children. Some of the defenders hold large hide shields that cover their bodies as they shoot arrows at the attacking forces. I now discuss specific sites that I believe contain Apache rock art. Next I compare them to the body of knowledge that has been developed about Athapaskan rock art elsewhere in the intermountain West.

168

L AWRENCE L. LOENDORF — THUNDER

AND

HERDS

FIGURE 6.1. The Segesser I hide painting. Note the rock wall fortification protecting the women and children and the large buffalo hide shields used by the defending Apache. Courtesy of the New Mexico Palace of the Governors (MNM/DCA), Neg. No. 149778.

THE SUE SITE The simple name “Sue” has more resonance than the three letters used to spell it would suggest. Not only does it figure in the name of a well-known Johnny Cash song, it has been attached to the large Tyrannosaurus Rex fossil now on display at Chicago’s Field Museum, which was named for its discoverer. The Sue site in this discussion was also named for its discoverer, an archaeologist who found it in the summer of 1983 while she was engaged in a survey project at the Piñon Canyon Maneuver Site. Located within a long, high-walled rockshelter opening to the west on Van Bremer Arroyo, the site’s main sheltered area measures nearly 15 meters along its north-south axis and 5 meters from east to west. Normal stream flow and flooding in the arroyo have exposed major outcrops of Dakota sandstone which form a cliff wall that is approximately 20 meters high in the site area. Along the wall, one group of petroglyphs is found at ground level, and others are distributed 1.5 to 2 meters above the present ground surface. Two significant excavations in the deposits at the site have revealed a series of stratified cultural layers that, although somewhat disturbed by the arroyo’s cutand-fill sequences, remain sufficiently intact for the relatively continuous use of

6 — ROCK ART

OF THE

PROTOHISTORIC PERIOD

169

the site from the Late Archaic to Historic Periods to be visible (Figure 6.2). The lowest levels, at nearly 3 meters below the surface, have been poorly sampled, with the result that the Late Archaic use of the site is not well defined. The primary cultural remains were deposited during the Developmental Period and include slablined hearths, ground stone implements, Scallorn and other types of cornernotched projectile points, and a variety of other chipped stone tools. Upright slabs of sandstone within the borders of the excavation units are probably the remnants

FIGURE 6.2. A contour map of the Sue site showing the locations of the excavation units and the main petroglyph panel. Map by David D. Kuehn, who supervised the University of North Dakota excavations (Loendorf and Kuehn 1991).

170

L AWRENCE L. LOENDORF — THUNDER

AND

HERDS

of house walls, but because they were found at a depth of more than a meter, they were not exposed and their configuration remains unclear. Radiocarbon dates for the lower levels, which constitute approximately three-quarters of the cave deposits, range from A.D. 380 to 980.9 The thick Developmental layers were overlain by a sparse level that contained small side-notched projectile points and various kinds of other chipped stone tools. Also included in the layer was a large, dark stain of mixed charcoal and ash that produced a radiocarbon date of A.D 1230. Above this layer, the deposit was essentially sterile until the uppermost 10 to 15 centimeters, which contained several shallow hearths, an unnotched projectile point, and a single thin-walled, undecorated pottery sherd. A radiocarbon date of A.D. 1520 to 1640 was obtained on material from this level.10 Faunal and floral remains in the stratified deposits at the Sue site are of considerable interest. There is no evidence of any cultigens in any level. The ground stone tools found at the site appear to have been used primarily to process sunflowers and chenopods, but the presence of cactus spines, hackberry seeds, wild onions, and gromwell seeds suggests these species were also included in the diet. Faunal remains included bison bones in both the lowest levels dating to the Late Archaic and in the uppermost levels. The thick layers dated between A.D. 380 and 980 contained the remains of jackrabbits, cottontails, prairie dogs, and jumping mice, but there is little evidence of larger mammals such as deer, antelope, bighorn sheep, or bison. The large quantity of freshwater mussel and crayfish remains suggests the site was used during a warm season. The combination of cultural materials in the Sue site deposits indicates that it was occupied by small groups of hunters and gatherers on an intermittent basis, perhaps in the fall and spring.

Rock Art at the Sue Site Thirteen petroglyph panels occur at the site, nearly three-quarters of which face west—the primary orientation of the sandstone wall—toward the arroyo.11 The remaining one-quarter face the interior of the rockshelter, placed by the makers on the relatively small indentations and protuberances in the wall, which makes them visible only from within the site and not from the arroyo (Figure 6.3). Several different processes were used to create the images: 49 percent of the petroglyphs were made using a stipple-pecked technique, 45 percent were solidly pecked, and those remaining were produced using a combination of these two techniques. The high percentage of stipple-pecked figures at the site is significant, since solidly pecked petroglyphs are the norm in the region. Several characteristics of the Sue site petroglyphs call attention to differences from rock art of earlier time periods. First, and perhaps most importantly, the figures were defined using a pecked outline technique, unlike the solidly pecked Purgatoire Pecked petroglyphs. Another difference is that 30 percent of the pet-

6 — ROCK ART

OF THE

PROTOHISTORIC PERIOD

171

FIGURE 6.3. The main petroglyph panel at the Sue site. Note the single shield-bearing warrior in the lower center, the outline-pecked and totally-pecked anthropomorphs, and the small turtle in the upper left. Scale drawing by Janet Lever.

roglyphs consist of horned heads which occur either as part of anthropomorphic figures or as independent images. The bodies of horned anthropomorphs are outlined by linear pecking, and the figures have upraised arms that end in fingers. Visual emphasis is clearly on the heads, both through the depiction of eyes and mouths and the prominence of horns that protrude upward from the head in both straight and curved lines. One figure appears to represent a shield-bearing warrior whose weapon is shown protruding from behind the shield. Depictions of bison represent another departure from past traditions. Although one bison figure is solidly pecked, the others—shown in profile with elongated rectangular bodies, long tails, and horned heads—are only outlined (Figure 6.4). One figure is a very good likeness of a bison, but the others are more crudely made. The incised heart line of one bison figure connects its mouth to the interior of its body. The head of another bison, whose eyes are defined, is turned to face the viewer. In this position the head is similar to the images of nearby horned heads. Sometimes an apparently insignificant attribute of a petroglyph figure implicates the existence of an important relationship between it and other figures in the region. One such feature—a short line intersecting the legs of one bison immediately above its downward-pointing, U-shaped hooves—occurs on the petroglyphs of other quadruped species at nearby sites (Figure 6.5). This convention is recognized as a representation of a dew claw. The presence of dew claws in several different contexts suggests that the inventory of protohistoric Apache pecked animals may be expanded, given sufficient study. Although a solidly pecked turtle petroglyph is the only other animal figure at the Sue site, abstract figures are present and include an asterisk-like motif, a star-like

172

L AWRENCE L. LOENDORF — THUNDER

AND

HERDS

FIGURE 6.4. A bison petroglyph at the Sue site. Photograph by Peter Halter for the United States Army.

FIGURE 6.5. A bison at the Sue site that has a short line intersecting the legs immediately above the inverted U-shaped hooves. The horned head to the right of the bison is another common figure at the site. Photograph by Peter Halter for the United States Army.

cross, and a rayed circle connected by a long line to another circle. Groups of evenly spaced dots occur in various locations at the site. Three of these are circles—with diameters measuring 32 centimeters—of such uniform roundness as to be unusual.

Implications of the Rock Art at the Sue Site Petroglyphs at the Sue site differ from Purgatoire pecked figures in two important ways: as noted previously, stipple-pecking is the production technique of

6 — ROCK ART

OF THE

PROTOHISTORIC PERIOD

173

choice and, with the exception of one or two figures that may be part of an earlier tradition, pecked outlines rather than infilled forms define the figures. Several facts indicate that the Sue site petroglyphs were made by Athapaskanspeaking Apache groups. First, there is a distinct similarity between figures at the Sue site and those in New Mexico that are included in the rock art category referred to as “Rio Grande style,” which occurred after A.D. 1300–1400. Petroglyphs at the Sue site are not identical to Rio Grande–style figures, but they are clearly related or influenced by that style. I expand further on this important distinction when I discuss the occurrence of regional sites whose petroglyphs appear to be genuine expressions of Rio Grande style. A second reason to identify the Sue site petroglyphs as Apache is the presence of a heart line motif in one of the bison. The heart line does not occur on any Purgatoire Pecked quadrupeds, and it is generally thought to be an attribute of rock art from the Plains that was introduced by Athapaskans into the American Southwest.12

THE S TONE S TRUCTURE SITE It is likely that through ongoing research, archaeologists will be able to make increasingly secure statements about the role played by rock art in the lifeways of the groups that created it. At present, however, the function of most rock art sites remains to be discovered. For that reason, and as I noted in my discussion of the Petroform site in Chapter 4, the presence of associated structures or other cultural remains at rock art sites expands our opportunities for learning about the past. The features at the Stone Structure site, which is located at a high point on the Piñon Canyon Hogback, about a kilometer from the Petroform site, therefore add another dimension to the investigation of the site’s rock art. A comparison of the structures at the Petroform and Stone Structure sites reveals significant differences in their construction. At the former site, structures were made by stacking basalt blocks in circles and other shapes on a flat bench near the Hogback. The rock features at the Stone Structure site, however, are part of the dike itself and occur in three different configurations. Two of these involved the removal of fractured basalt blocks from the dike to create threesided “nests” or protected areas within the dike wall. One such cavity is 3 meters wide, 1.37 meters deep, and 0.6 meter high, and the other is 1.7 meters wide, 1.5 meters deep, and 0.77 meter high. The third structure was made by stacking basalt blocks to form a three-sided wall attached to the dike. Its long axis is almost directly east-west, the side attached to the dike wall is 3.1 meters long, and its width—or the distance that the structure protrudes from the dike—is 1 meter. The average height of the wall is 0.5 meter. Another important difference between the two sites is that the 22 petroglyph panels at the Stone Structure site are directly associated with the stone features

174

L AWRENCE L. LOENDORF — THUNDER

AND

HERDS

embedded in the Hogback formation. Quadrupeds, which are the dominant figures in the panels, are fully pecked, but they differ from Purgatoire Pecked quadrupeds in the attention given to detail in the depiction of such body parts as horned heads, hooves, and tails. Several panels of bison include groups composed of animals of different sizes (Figure 6.6). Human figures are not as prevalent as quadrupeds, but they constitute an important part of the suite of images at the site. Some are full-frontal figures whose inverted, V-shaped legs are attached to the base of a straight, stick-like body and whose arms are represented by a straight, horizontal line crossing the body at the chest. Heads are indicated by slightly larger, rounded areas at the top of the body. One prominent anthropomorph is more complete. Its pecked rectangular body has straight legs connected to the horizontal line forming the body’s base. Its arms are upraised, and the figure appears to be holding something in one hand. It also appears to be wearing a cross-shaped object on its head. A range of geometric figures is found at the site. Several images have shapes indicating that they may represent plants, and one abstract oval figure resembles a turtle. Grid-like patterns and other abstract forms are present, but the deeply pecked, meandering lines and circular forms typical of the Archaic Period are absent. All of the figures at the Stone Structure site may not be contemporaneous. The heavy varnish on two or three very old-looking quadrupeds nearly obliterates their form, which probably means that the figures date to the Late Archaic or Developmental Period. The rudimentary form of some of the stick-like human figures may indicate that they date to the Diversification Period. The overwhelming majority of the figures, however, are more recent in age, an assessment supported by three types of evidence. First, well-crafted representations of bison, which are absent at Purgatoire Pecked sites, are found at the site. Second, at another Hogback site (discussed below), bison figures are superimposed on older quadrupeds representing deer or antelope. And finally, the results of cationratio dating of the bison petroglyphs, which with one exception are in a 350- to

FIGURE 6.6. Representative bison and deer from the Stone Structure site. Note the short line crossing the legs of the deer above the hooves. Illustration based on a tracing by Janet Lever.

6 — ROCK ART

OF THE

PROTOHISTORIC PERIOD

175

400-year range—A.D. 1550 and 1600—are uniformly more recent than the dates obtained for Purgatoire Pecked quadrupeds.

Possible Activities at the Stone Structure Site What can the properties of the rock constructions and their association with rock art tell us about the activities that might have taken place at the site in the past? The first fact of interest is that these structures are old. The rock floors are heavily varnished from exposure to the elements, and lichen is growing on the exposed surfaces and between the rock walls. Despite their obvious antiquity, however, the structures contain no evidence that they were used as habitations. No bedrock metates and no artifacts of any kind—either chipped or ground stone objects or ceramics—were recovered from the site. At least four models attempt to explain the function of sites with stone structures. Inherent in the first two is an assumption that the rock art was created during the course of activities that constitute the site’s primary purpose. The remaining two models emphasize that the creation of rock art was a mechanism for achieving the goals of persons using the site. For example, based on their position in the surrounding landscape, it has been suggested that such sites might be hunting blinds. For several reasons, however, such a proposition is unrealistic. First, an exposed area high on a stone dike—like that at the Stone Structure site—is not a practical place to situate a hunting blind. Although visibility is excellent in all directions, if the hunting strategy involved driving prey animals into a catchment area below the Hogback, there would be no reason for hunters to crouch behind the structures’ low walls. A much more likely scenario is that a hunter would hide on the opposite side of the dike, where observation of prey from a concealed vantage point is possible. Furthermore, ethnographic studies report that hunters keep busy by sharpening their tools or making arrows while they wait for game to appear,13 and these activities produce lithic debris and other kinds of detritus not found at the Stone Structure site. A second proposition is that such sites were observation posts where sentries waited and watched for either game or enemies. Although more probable than the hunting blind hypothesis, this scenario is unlikely for one of the same reasons— the absence of any associated artifacts. For purposes of comparison, it would be helpful if there were a record of the excavation of at least one prehistoric sentry post and a description of any recovered artifacts, but to my knowledge none exists. Until comparative information becomes available, knowledge of human behavior and doubts about the likelihood of persons spending long hours watching their surroundings without leaving any evidence of their activities will have to substitute for facts of relevance. A third idea is that sites with stone structures represent places where vision quests were undertaken. This suggestion has merit because of the resemblance of the

176

L AWRENCE L. LOENDORF — THUNDER

AND

HERDS

features themselves to the stone structures on the northern Plains that are known to be vision quest fasting beds.14 Ethnographies of the Crow Indians, for instance, describe how individuals on vision quests built stone structures and used them as places where they fasted in anticipation of spiritual transformation. From an explanatory standpoint, however, this proposition is unhelpful because the vision quest was not part of the spiritual practice of southern Plains tribes, so the features at the Stone Structure site cannot be explained by reference to this practice. Instead, as reported by Morris Opler, groups like the Apache had guardian spirit helpers. A basic tenet of Mescalero Apache religion is that supernatural power and guidance can be obtained from various animals, plants, natural forces, and guardian deities. In a dream or vision the power source appears to the Mescalero Apache first in its customary form (as an animal or a plant, for instance); then it usually assumes human guise, describes, and teaches the songs, prayers, and other particulars of the ceremony if the Apache is willing to learn and serve.15 From Opler’s account it is clear that instead of engaging in a vision quest ritual like that of northern Plains groups, which involved seclusion and fasting, an Apache supplicant received power and guardian spirits in a dream. As Opler describes, “one day a person may have ‘something speak to him.’ It may come in a dream; it may be when he is with a crowd of his fellows.”16 This account strongly suggests that the features at the Stone Structure site are not the remains of vision quest structures like those found farther north. I have used ethnographic sources to question the validity of the proposed similarity between the features at the Stone Structure site and vision quest sites on the northern Plains, but these sources also include information making it possible to assert that this site and others like it represent Apache prayer stations. I expand on this possibility, the last of the four models I discuss, in greater detail in a subsequent section of this chapter, but my argument is similar to the one suggesting that such sites were created and used as locations where vision quests were pursued. Part of my reasoning is based on the fact that the Stone Structure site is not unique. A basalt dike, located on a Colorado ranch near the border with New Mexico and referred to as the Smith site, has features similar to those on the Hogback in Piñon Canyon (Figure 6.7). Significant petroglyph sites are located along its 10-kilometer length, one of which includes anthropomorphic figures, bison, and other animals that are directly associated with stone structures. The latter are approximately 2 meters long, 1 meter wide, and 1 meter deep. Unlike the structures situated on the sides of the Hogback in Piñon Canyon, these were made by hollowing out areas within the dike itself. The similarities between the Stone Structure and Smith sites, and others likely to be in the region, indicate that they are the work of the same cultural group, and their placement relates to the belief—later recorded in ethnographic sources—that basalt dikes were the abode of Apache mountain spirits (Figure 6.8).

6 — ROCK ART

OF THE

PROTOHISTORIC PERIOD

177

FIGURE 6.7. A pit str ucture in the Smith dike that is very similar to features at the Stone Structure site. Photograph by Lawrence Loendorf.

FIGURE 6.8. An anthropomorph at the Smith dike that appears to represent a lightning gans. Illustration by Bonita Newman.

178

L AWRENCE L. LOENDORF — THUNDER

AND

HERDS

COMPARISON AND DIFFERENTIATION In this section I present a comparison of the rock art motifs at sites in the region that appear to represent Apache rock art. First I discuss the representation of animal and human figures and then analyze figures whose appearance suggests they are depictions of very powerful mountain spirits. I then examine the contribution that a pollen analysis makes to a determination of the ethnic identity of the creators of the Piñon Canyon rock art record. Finally, the results of dating techniques used to develop a chronology of the rock art record of the area are evaluated. Bison petroglyphs are the most distinctive animal figures in the region during the Protohistoric Period. In the preceding Developmental and Diversification Periods, some ambiguous animal figures, such as the quadrupeds at the Zookeeper and other sites, may represent bison, but their ovoid bodies and stiff legs are not definitive and could just as easily be depictions of bighorn sheep. In contrast, the rock art bison that I believe to be Apache in origin are excellently rendered and leave no question of the maker’s intent. Although a few examples are completely pecked, most of the figures are outlined and have interior features, such as ribs or heart lines. In nearly every instance, the figures have well-defined hooves that are anatomically accurate or are shown with dew claws. At several sites in the region, the dew claws also occur on well-made images of deer found at sites with bison. Although figures of deer are not as prevalent as bison, their co-occurrence suggests that both quadrupeds are examples of Apache rock art (see Figure 6.6). Heart lines are an important component of the regional Apache rock art. They are found on the bison figures at the Sue and Smith sites, as well as on the compelling bison petroglyph at the Red Guns site, near La Junta, Colorado. The latter outlined figure has an open mouth, through which a heart line has been pecked into the interior of its body. A Navajo bison petroglyph from a site in the Dinetah is amazingly similar in form, leaving little doubt that both figures were produced by the same cultural tradition (Figure 6.9). The heart line is a complicated motif, however, and because it was widely used elsewhere on the Plains, its simple presence is not sufficient evidence to ensure that a figure is Apache. In this context, production technique is the attribute that allows archaeologists to distinguish between Apache and non-Apache rock art. When a number of historic and possibly older incised animal figures with heart lines—in western Oklahoma and at Picture Canyon in southeastern Colorado—are examined in light of this criterion, it becomes apparent that they are not part of a pecked Apache tradition. Pecked images of turtles—like those at the Red Guns site, the Sue site, and Smith dike—also figure significantly in Apache rock art. Among contemporary Navajo and Apache peoples, images of turtles are regarded as war-related, and in

6 — ROCK ART

OF THE

PROTOHISTORIC PERIOD

179

some accounts, because of the turtle’s hard shell, they are interpreted as symbols of Athapaskan shields.17 In contemporary Navajo sand paintings, turtles are portrayed both realistically and as abstract forms, a fact that has proved helpful in solving an interpretive problem presented by a stylized, turtle-like petroglyph at the Stone Structure site (Figure 6.10). When the image in question was compared

FIGURE 6.9. A bison with heart line from the Dinetah compared with a similar bison from southeastern Colorado. Illustrations reworked from a photograph of the Colorado bison and a line drawing by Klaus Wellman of the Dinetah example.

FIGURE 6.10. On the left is a turtle-like figure at the Stone Structure site. The figure on the right is a sand painting of a turtle. The Stone Structure site image is an illustration by Elaine Nimmo from a field drawing by Claudette Piper. The sand painting illustration is by Bonita Newman.

180

L AWRENCE L. LOENDORF — THUNDER

AND

HERDS

with a Navajo sand painting of a turtle, the strong resemblance between the two images was sufficient to place the petroglyph in the turtle category. In the search for evidence linking rock art images to an ethnic group such as the Apache, historical records are valuable sources of information about material objects depicted in petroglyph panels. For example, among researchers it is generally agreed that the sinew-backed bow was one of the distinctive items of hunting technology that immigrating Athapaskans brought with them to the Southwest.18 Sinew-backed or composite bows have curved backs and are easily distinguished from so-called self bows—which are made from a single piece of wood or horn and have D-shaped backs—which were used throughout the Southwest and on the southern Plains. A recognizable rock art image of a sinewbacked bow is held by a warrior figure on the Piñon Canyon Hogback (Figure 6.11). This figure, wearing a horned headdress, has an outlined rather than infilled body, digitate hands, and plantigrade feet. The open body is a consistent feature of human figures in Apache rock art elsewhere in southeastern Colorado. The representation of human figures in Apache rock art is not uniform, however. Like bison, some human forms are made by simply pecking the outline of the figure, while other images have completely pecked, usually rectangular bodies. It is relatively common for some portion of the figure to be left in an unaltered state, its rock varnish coating still intact. Legs descend from the body

F I G U R E 6.11. An example of an Apache figure carrying a sinewbacked bow. Illustration by Chris Alford based on a photograph by Lawrence Loendorf.

6 — ROCK ART

OF THE

PROTOHISTORIC PERIOD

181

as vertical lines, and the arms are often upright, ending in digitate hands. Some heads on the outlined figures have eyes and a mouth. Horned headgear is the most common adornment associated with these figures, although some heads support cross-like shapes or candelabra forms, and other figures have no head appendages at all (Figure 6.12). Human figures are variable in size as well. Most are 20 to 30 centimeters tall, although a few, such as the large figure that dominates a panel at the Red Guns site, are nearly 2 meters tall. Figures of humans are often arranged in groups and are less frequently associated with animals than the anthropomorphs at Purgatoire Pecked-II sites. Several examples, however, appear to be accompanied by dogs, an association supported by descriptions in Spanish chronicles of Apache individuals and their canine companions. Occasionally a human figure, such as the anthropomorph I described earlier that is carrying a sinew-backed bow, is depicted holding an object. There are a few figures that may be gripping shields, which they hold in front of them to cover their bodies. Unambiguous shield-bearing warriors occur across the northern Plains and in the Dinetah, but these are quite rare in southeastern Colorado. Before discussing this subject in greater detail, however, I need to introduce the most dramatic, identifiably Apache figures, the gans. Ethnographic sources record that these figures, also called gan dancers or crown dancers, represent many different kinds of Apache mountain spirits. Although gans are smaller than humans, the resemblance to them is strong. Gan

FIGURE 6.12. Petroglyphs representative of the Apache anthropomorphs in the region. The bodies of many figures are outlined-pecked, but some examples, like the figure in the center, are infilled. Horned heads, like those from the Sue site on the right, are common, but the branching headgear of the center figure is rare. The figure on the left, associated with a small dog, is approximately 40 centimeters tall. Drawings made by Elaine Nimmo from photographs by Lawrence Loendorf.

182

L AWRENCE L. LOENDORF — THUNDER

AND

HERDS

spirits wear a G-string and high-top moccasins, and they have candelabra-like headdresses attached to the tops of their heads (Figure 6.13). Some carry spruce branches in their hands, while others have knives and are reputed to cut to pieces any humans who presume to touch them.19 Gans make their homes in rock formations, which they enter through caves or small holes. Features of their secret world include miniature landscapes with trees, grass, and water, and the sites where they occur are considered holy places. Anthropologist Grenville Goodwin, who lived with the San Carlos Apache for nearly a decade, recorded that there was no difficulty recognizing rock drawings made by a gan: About one day’s journey west of Zuni there is a place on the bluff of a hill where the gan people have pecked pictures on the cliff higher than you can reach. They have pictures of themselves there. You can tell when the gan have done it and when the [shamans] have done it, because only the former make pictures that are up above your reach on the cliff. The latter usually make them in caves.20 Apache shamans hoping to acquire power in a trance or vision would sleep near gan sites.21 Although each shaman’s experience was different, there were many commonalities. Gan spirits would appear to a man who was praying or sleeping near a rock art site with gan petroglyphs and bid the supplicant to follow them into the rock. Once inside, the shaman would have to overcome four obstacles and then pass through separate chambers, where animals and supernatural beings of all kinds offered a variety of powers. Following the advice of the gans, the supplicant would decline all offers and proceed to the final chamber, where the gans would be waiting for him. There the shaman would accept the powers presented by the gans and spend “four nights learning the details of the ceremony and the designs to be painted on the persons and paraphernalia of the dancers. His guide conducts him back to the door once more and then the Apache finds himself awake at the place where he lay down to rest.”22

FIGURE 6.13. Gan dancers from the Piñon Canyon Hogback. Drawing by Hanna Hinchman based on a tracing by Kerry Hackett.

6 — ROCK ART

OF THE

PROTOHISTORIC PERIOD

183

Once a shaman had acquired power, he would be permitted to organize and officiate at gan dances and ceremonies for his people. These events were often associated with a particular healing power, such as bear medicine, which was used to heal wounds or other illnesses caused by bears. Individuals visiting the gans’ sacred homes to pray and seek guidance frequently brought pollen from the cattail plant (Typha latifolia) as an offering. The Apache used cattail pollen in curing ceremonies and all other encounters with the supernatural, and it was known to be especially effective during episodes of lightning and thunder, during which a man would scatter pollen in the direction of the approaching storm.23 In an effort to learn whether rock art sites with gan-like figures were associated with the activities of Apache shamans, I had several soil samples analyzed from sediments within the rock features on the Stone Structure site and found cattail pollen present.24 Based on the facts that cattail plants do not disperse their pollen very far from where they grow, and that there are no stands of cattails near the Structure site, where the soil samples were taken, researchers concluded that the pollen was transported to the site by humans. I and others believe that Apache shamans considered the gan petroglyph sites on the southern High Plains basalt dikes to be the homes of these mountain spirits and visited the sites to seek the special powers offered by these supernatural deities. The stone structures on these sites are, therefore, vision quest locations where individuals waited for visits from the mountain spirits.

Shield-bearing Warriors The near absence of shield-bearing warrior figures at rock art sites in the central High Plains contrasts sharply with the frequency of these figures (as well as images of stand-alone shields) on the northwestern Plains, where they are produced by a half-dozen different techniques. One particular type—referred to as Castle Gardens style—is thought to be linked to Athapaskans.25 Several techniques and at least three steps were required to make Castle Gardens figures. First, an abrader was used to smooth the surface of a sandstone wall, after which the outline of a shield and any intended decoration were incised into the prepared surface. Three or four colors of paint were then applied to different sections of the design. Although there is considerable variation in the figures’ size, most are approximately 75 centimeters tall. Now badly faded, when the painted colors were fresh, these shield figures must have been stunning. An important variable linking Castle Gardens figures to Athapaskan groups is their age. During an excavation at the base of a panel of Castle Gardens–style figures at the Valley of the Shields site in south-central Montana, sandstone abrading tools to which paint still adhered were recovered from the deposits. The tools were found in association with a hearth containing charcoal that was dated to A.D. 1100, which is within the time period during which Athapaskans are believed to have been present in Montana and Wyoming.26

184

L AWRENCE L. LOENDORF — THUNDER

AND

HERDS

The geographical distribution of Castle Gardens shield figures is also significant. They have been identified at 15, and possibly 16, sites in the 600 or so miles between the Musselshell River in central Montana and localities in northcentral Colorado. One common characteristic of the rock art images at most of these sites is the presence of green pigment—an unusual color in the pictographs of Montana and Wyoming—on many of the Castle Gardens–style shields. An important and direct link between the sites comes from the fact that the ingredient responsible for the paint’s green color is fuchsite, or green mica.27 The most compelling evidence of an association between Castle Gardens– style shield figures and Athapaskan rock art is found in the Dinetah itself. Numerous shields and shield-bearing warriors in the Dinetah are attributable to Athapaskan-speaking peoples, and many of these figures were created by the same multistep process—abrasion of the rock surface prior to incising the designs and filling them with paint—that produced the Castle Gardens shield figures. Athapaskan rock art figures also have horned headdresses and other design elements that link them to the Castle Gardens–style figures. On the other hand, only a dozen or so known rock art shields have been found on the central High Plains. Six shield-bearing warriors, painted in white, are shown walking single file on a rock art panel at a site to the east of the Laramie Mountains. Two eroded figures in the line appear to be carrying packs or burden baskets on their backs.28 These shields figures, which differ in significant ways from Castle Gardens–style figures, probably date to the 1600s, which makes them 400 to 500 years younger than the Castle Gardens figures. Photographs of four faded and badly eroded painted shields in southeastern Colorado’s Picture Canyon indicate that red and buff pigments were used to paint the figures. The interior areas of the two most visible examples were painted with a buff-colored wash, except in those areas where designs were created by leaving the sandstone sediments uncovered. Inverted triangles occur on the face of one figure, and a red plant-like design has been superimposed over a dot pattern in a white area on the other figure. Because the shields are multicolored, they resemble Castle Gardens–style figures, but the rock surfaces were not prepared prior to painting, and the designs were not incised within the shields. These figures may be Apache in origin, but they could as easily represent the work of another historic Plains tribe.

Rio Grande Style Southwestern rock art researchers have identified a complex of images and themes—known collectively as Rio Grande style—that occurred primarily in the rock art of New Mexico between A.D. 1350 and the beginning of the historic period in A.D.1725. Rio Grande style is closely linked to ancestral Puebloan groups. Since shields and shield-bearing warrior figures are included in the style,

6 — ROCK ART

OF THE

PROTOHISTORIC PERIOD

185

when these figures are found in the High Plains research area, the following questions become relevant: What is their provenance, and to what stylistic tradition do they belong—Apache or Pueblo? The largest numbers of shields and shield warrior figures in the region occur at a site in the Picket Wire Canyon bottomlands, near the mouth of Withers Canyon.29 A significant characteristic of this site, referred to as the Battle site, is the presence of several bas-relief figures, so-called because the dark varnish surrounding the figures has been removed (Figure 6.14). One figure consists of a shield on whose surface the bust of a human figure has been depicted. The figure’s neck is embellished with vertical, striped lines, and its round head, on which there is a very distinctive spiked headdress, contains eyes and a nose. On another shield, only the head of a figure wearing a spiked headdress appears (Figure 6.15). The same petroglyph panel includes images of warriors with rectangular bodies, bows, arrows, and hatchets, as well as a human hand in basrelief. Elsewhere in the Picket Wire, a fully pecked shield warrior with bent knees appears to be walking in close association with five dogs that have pointed ears and long tails. The juxtaposition of dogs and shield-bearing warriors is rare because dogs, who had an important role as village sentries, did not usually accompany Indians on their raiding parties. The Spanish use of dogs in war is well known, but to my knowledge they were not present on any of the punitive expeditions led by Spaniards into El Cuartelejo. Even more out of place in the rock art inventory at the site is the distinctly drawn petroglyph of a corn plant.

FIGURE 6.14. Rio Grande–style warriors in profile view carrying bows and hatchets. Photograph by Lawrence Loendorf.

186

L AWRENCE L. LOENDORF — THUNDER

AND

HERDS

FIGURE 6.15. A Rio Grande–style figure in bas-relief that may adorn a shield. Photograph by Lawrence Loendorf.

A pecked shield featuring the bust of a human figure, similar to the one just described and wearing a spiked headdress, is found about 30 kilometers from the Picket Wire Canyon site. Made by pecking out the image within the circular shield, this figure is stylistically similar to those at the Battle site. These two sites, and possibly others on the central High Plains, contain primarily war-related images and appear to represent examples of Rio Grande–style rock art. The images depicted at these sites may represent expeditions onto the High Plains by Pueblo war parties or, alternatively, they may reflect the adaptation of Rio Grande–style subjects and images by Apache peoples.

Dating Apache Rock Art A series of nine cation-ratio dates on petroglyph varnish samples—from a bison image at the Sue site, the human figure holding a sinew-backed bow, several gan figures, and the well-made deer at the Stone Structure site—range from less than 300 years to approximately 400 years before the present.30 These dates suggest

6 — ROCK ART

OF THE

PROTOHISTORIC PERIOD

187

that the figures are more recent than Purgatoire Pecked-II petroglyphs, which means they are likely to postdate A.D. 1450. The single radiocarbon date of 370 ± 60 B.P. on a sample from the uppermost cultural layer at the Sue site may be important for establishing an actual time frame for Apache use of the rockshelter.31 This date has a calibrated age, with 95 percent probability, that ranges from A.D. 1440 to 1644. Artifacts from the same level included polished Ocate Micaceous (Apache) ceramics and an unnotched, triangular arrow point. If additional radiocarbon dates had been obtained on material from this layer, it may have been possible to place the site more precisely within the 200-year time period. In New Mexico, however, Ocate Micaceous ceramics are dated between A.D. 1600 and 1750; so by combining the radiocarbon date range and the ceramic data, it seems likely that the site dates between A.D. 1600 and 1644. Other aspects of the rock art itself, such as the superpositioning of a wellcrafted bison at the Smith dike site over a petroglyph figure from the Developmental or Diversification Period (Figure 6.16), supports the sequence proposed in this chapter. Superpositioning provides reliable, if relative, information about the age of rock art figures, as is demonstrated by another example in which incised horses and riders are superimposed over Apache rock art figures at several sites in the region. By linking the results of superpositioning at these sites, however, we can say that Apache petroglyphs postdate Purgatoire II figures and predate rock art featuring horses and riders. Chronological estimates can also be inferred from the amount of varnish present on those petroglyphs believed to be Apache. There is very little evidence

FIGURE 6.16. The figure of a bison superimposed over an older Apishapa-age pecked quadruped. Illustration by Terry Moody.

188

L AWRENCE L. LOENDORF — THUNDER

AND

HERDS

that any darkening of pecked surfaces has occurred between the sixteenth century and the present and, not unexpectedly, the varnish covering Apache rock art is invariably light. At many sites, in fact, the varnish is so minimal that if an archaeologist were to try to estimate the relative age of a petroglyph based solely on the appearance of the varnish, Apache figures would be judged to be contemporaneous with initials and dates from the Historic Period. Although the group of dates I have presented here does not include any single example that would definitively establish the time period when groups of Apache were present in the High Plains region, taken together they point to an initial Apache presence sometime around A.D. 1600. Chris Zier and his colleagues at Centennial Archaeology reached the same conclusion after an examination of all the dated protohistoric sites in the region.32 They concluded that there is very little archaeological evidence for an Apache habitation on the High Plains in the 1500s.

CONCLUSIONS I now believe that a solid case has been made for the identification of an Apache rock art tradition in the Purgatoire River region. Like other researchers trying to identify Apache rock art in southeastern Colorado, I had previously placed this tradition within the broader Rio Grande rock art style.33 In this book, however, I am proposing a new category—Apache style—in order to give the classification regional autonomy. The petroglyphs in this newly identified Apache style are characterized by pecked outlines and frequent use of a stipple-pecking technique. The most distinctive animals are well-made bison that often have heart lines or ribs. Bison and quadrupeds such as deer and elk frequently have either a dew claw or some other extra hoof feature. Both outlined and solidly pecked turtles are present and have obvious legs and tails. At least one turtle resembles a sand painting turtle. Human figures often carry sinew-back bows. Shield-bearing warrior figures are rare, and those present have large, circular pecked shields with little decoration. Humans are often shown with horned headdresses. Painted human figures have narrow waists and very small Y-shaped heads that may indicate the presence of horned headdresses. Some anthropomorphs appear to be wearing cross-shaped or candelabra-like headgear. I believe these figures represent the gans who played such a special role in the Apache lifeway. Nonrepresentational designs include detached starbursts as well as ones connected to other circular forms. At one location an arrow penetrates a starburst. Arrow points unconnected to arrow shafts are another common motif. High Plains Apache petroglyphs occur primarily between A.D. 1600 and 1725, although some examples may be slightly earlier or later. At present the

6 — ROCK ART

OF THE

PROTOHISTORIC PERIOD

189

style is found at sites concentrated in the Purgatoire River drainage, but some petroglyphs have been found as far north as the Arkansas River and into the Cimarron drainage to the south.

NOTES 1 The Compact Edition of the Oxford English Dictionary 1971: s.v. “proto-.” 2 The definition of protohistory appears in the first sentence of the section with that name in the

online encyclopedia Wikipedia. It can be accessed by using the following URL: http://en/ wikipedia.org/wiki/Protohistory. 3 Matson and Magne 2007. In the final chapter of this book I discuss in greater detail the implications of these authors’ ideas. 4 Opler 1983:382; Towner 2003:194–207; Gunnerson and Gunnerson 1971; C. Schaafsma 2002. 5 Gill 1983:502; Yazzie 1971. 6 Towner 2003. 7 The geographical extent of the area to which the term El Cuartelejo is applied comes from Carrillo (1999) as cited in Kalasz et al. 1999:251. 8 Hotz 1970. 9 Andrefsky et al. 1990:272. 10 Loendorf and Kuehn 1991:77. 11 Loendorf 1989:281. 12 P. Schaafsma 1980:307; Baldwin 1997. 13 In a series of articles in the 1970s, subsequently collected in Working at Archaeology (1983), Lewis Binford reported extensively on Nunamiut Eskimo tool maintenance activities at hunting stands and their archaeological consequences. 14 Conner 1982. 15 Opler 1946:269 16 Opler 1935:67. 17 Clark 1983:97. 18 Brugge 1983; Baldwin 1997. 19 Goodwin field notes, June 24, 1936. 20 Goodwin field notes, June 24, 1936. 21 Opler 1965:312; Hoijer and Opler 1938. 22 Hoijer and Opler 1938:154. 23 Bourke 1892:500. 24 Scott-Cummings 1990. 25 Loendorf 1990, 2004; Francis and Loendorf 2002:143–44. 26 In the time since this book has been in press, I have obtained new radiocarbon ages for the cultural layer in the Valley of the Shields which suggest the A.D. 1100 date may be on "old wood" and therefore falsely old by several centuries. More research is underway to learn the true age of the Castle Gardens–style shield figures. 27 Newman and Loendorf 2005. 28 An illustration of this panel is used as the cover art for Colorado Prehistory: A Context for the Platte River Basin (Gilmore et al. 1999). 29 Reed and Horn 1995:166 30 Loendorf 1989:346; Loendorf and Kuehn 1991:266. 31 Loendorf and Kuehn 1991:57. 32 Kalasz et al. 1999:251. 33 Mitchell 2004a.

7 I N C I S E D A N D PA I N T E D R O C K A RT O F T H E H I S TO R I C PE R I O D

THE EQUINE COMEBACK he family tree of the horse (Equus caballus), although not thought of as a species indigenous to North America, first established its roots here between 50 and 60 million years ago. The species’ founding ancestor, Eohippus or Dawn Horse, was slightly larger than a lapdog and—similar to canids below the knee— had padded toes, four on each front foot and three on each back one. Remarkable morphological changes occurred in subsequent generations, as the tree sprouted as many as 15 different equine branches before stabilizing—two to three million years ago—into the horse species described in the Koran: “When God created the horse he said to the magnificent creature: I have made thee as no other. All the treasures of the earth lie between thy eyes.”1 In addition to populating the vast Pleistocene grasslands of the New World, herds of horses radiated northwest, crossing the Bering Strait during periods when an exposed land bridge provided access to Asia and beyond. The end of the last Ice Age reconstituted the Bering Sea, however, cutting off passage between Alaska and Siberia and leaving the horses of North America so maladapted to a continent in the throes of significant environmental change that within several thousand years the species was extinct. From a horse’s perspective, the fate of the equids that populated the Old World got off to a less than auspicious start. The fact that they were first considered prey animals is attested to by the presence of butchered horse bones in archaeological sites in many geographically distant places, from Botai, Kazakhstan, to the Lamb Spring site near the Littleton, Colorado. Evidence indicates

T

190

7 — HISTORIC PERIOD INCISED

AND

PAINTED ROCK ART

191

that equine domestication—a milestone for humans in terms of convenience but understandably less so for horses themselves—had occurred in the Ural steppes by 4000 B.P.2 Since that time the species has been selectively bred for characteristics that have enhanced its usefulness to humans involved in hunting, agriculture, sport, warfare, and transportation. The return of horses to the New World was involuntary; they were conveyed here in Spanish ships on voyages of exploration which resulted in the conquest of indigenous peoples and the imposition of European culture on two continents. Hernán Cortés was accompanied by 550 soldiers and 13 horses when he conquered Mexico in 1521, and by the mid-1600s, subsequent conquistadors had established colonies in a considerable portion of North and South America. It is not an exaggeration to say that an essential component of Spanish domination is the fact that the conquest was conducted largely on horseback. It is also not an exaggeration to add that the acquisition of horses by Indian tribes of western North America was an economically and socially transformative occurrence that has considerable relevance to the study of the rock art produced during the Historic Period on the central High Plains.

CHANGES IN THE E THNIC PLAYING FIELD OF THE P LAINS The chronological baseline against which the impact of large numbers of horses on Plains Indian lifeways will be measured begins sometime after A.D. 1400, when Caddoan-speaking groups moved from the central High Plains to locations farther east, where their subsistence focused primarily on maize, beans, and squash agriculture, supplemented by forays into the Plains to hunt deer and bison.3 After the establishment of Spanish settlements in the 1600s in what is now New Mexico, horses became available to the conquered Pueblo communities along the Rio Grande, to those Apache who had replaced Caddoan groups as the dominant presence on the central High Plains, and to the Pawnee, who intermittently exploited Plains faunal resources. Ownership of horses, although initially limited, transformed the ability of Indian hunters to access the large, grazing Plains ungulates, especially buffalo, that had been difficult for pedestrian hunters to pursue. Among those who were quick to take advantage of the new equine technology were Shoshone groups in Wyoming, who moved to the southern Plains. Once there, they formed alliances with their linguistic relatives, the Ute Indians, and successfully fought the Plains Apache for control of the region. These efforts ultimately forced the Apache to withdraw into the mountains of northern New Mexico and southern Colorado.4 The Ute were Numic speakers living in the mountains of Colorado and Utah, where hunting and fishing resources were so abundant that population

192

L AWRENCE L. LOENDORF — THUNDER

AND

HERDS

densities increased to one person per square mile in some localities.5 To cope with increasingly untenable demographic pressure, the Ute—who were organized into eleven bands, six of which had ranges in Colorado and five in Utah— joined the Shoshone with the acquisition of horses and other resources in mind. From their toehold on the Plains, Shoshone-Ute raiders positioned themselves closer to the newly established Spanish colonies in New and Old Mexico, attacking settlements and acquiring horses which they then took back to Wyoming to trade.6 The Pueblo Revolt of 1680 forced almost all Spanish colonists to temporarily evacuate their territory, and their precipitous retreat to safety in Mexico resulted in the abandonment of many of their horses and other animals. During a 12year absence, formerly Spanish livestock were redistributed among Indian groups, and the era of mounted, mobile competition for resources and territory began in earnest. By the early 1700s, as many as 10,000 Shoshone had left Wyoming, and this number may have increased considerably in the next 50 to 60 years.7 At some point in their armed entrepreneurial effort, the Shoshone-Ute raiders were given a new name by the Spanish—the Comanche—and although the origin of the name is obscure, the reality of the Comanche presence on the southern Plains is not. By A.D. 1740, the western Comanche had established themselves as middlemen in an extensive trade network that included transactions with French settlements along the Mississippi River, with the Spanish-Pueblo colonies along the upper Rio Grande, and with Indian groups as diverse as the Pawnee, Wichita, Kiowa, Kansas, Iowa, and eastern Comanche. Among the items purveyed were garden products, dried meat, bison hides, deerskins, and slaves, which were exchanged for horses, guns and ammunition, knives, kettles, iron arrow points, and a wide number of decorative items such as beads and pigments. The center of the trade network, which lasted until the 1830s establishment of Bent’s Fort near present-day La Junta, Colorado, was in the Comanche camps situated along the Arkansas River near its junction with the Purgatoire, in the heart of the central High Plains.8

GENERAL CHARACTERISTICS OF HISTORIC-ERA ROCK ART By the middle 1800s, the Arkansas River region was an ethnic crossroads, frequented by many tribal groups, including the Kiowa, Cheyenne, and Arapaho, and subject to attacks from horse raiders like the Crow, Arikara, and Shoshone from the northern Plains. Researchers believe that some, if not all, of these groups contributed to the rock art record of the southern High Plains, but identifying who is responsible for what is, not surprisingly, complicated by the area’s changing ethnic composition. I began this chapter by reviewing the contributions of that singular quadruped, the horse, to the dynamic changes on the High Plains beginning in the sixteenth century. It is now time to acknowledge the species’ role as a valuable chronological marker, whose presence in incised rock art enables researchers to date

7 — HISTORIC PERIOD INCISED

AND

PAINTED ROCK ART

193

the majority of those images to the Protohistoric and Historic Periods. Not only are incised horse images common, they are accompanied in many instances by guns, tipis, a variety of non-Indian human figures, horse tracks, and human footprints. Many of these figures are so lightly incised that they are sometimes said to have been scratched. In some sites, especially those with soft sandstone sediments, it is clear that the figures were incised with a sharp stone flake or a metal tool.9 A number of sites also depict horses and other figures that were drawn on the rock surface with lumps of charcoal. Although the medium used to produce the images is different, the figure inventory consists of the same mixture of horses, riders, tipis, guns, and weapons such as lances and shields. Red ocher was used to define some figures, either partially or totally, and a few of these figures are large, measuring more than a meter in their maximum dimension. The majority of painted images are, however, less than one-quarter that size. Incised and abraded grooves, often aligned in parallel rows, are common features of sites dated to the Historic Period. I will discuss these enigmatic figures, which resemble the petroglyph features found on boulders elsewhere on the Plains and are referred to as “ribstones,” later in this chapter. Following the precedent set in earlier chapters, I begin my detailed discussion of the complicated rock art record of the Historic era by focusing on specific sites that best typify its most pertinent characteristics.

OVERSIZE ROCK ART FIGURES Like other words ending in the suffix -size, oversize is a relative term. Although a 7-foot-tall member of the Los Angeles Lakers basketball team would be considered a player of average height when compared with his teammates, if he were standing next to the typical male archaeologist, he would be considered seriously oversize. To someone admiring the pictographs in the “Great Gallery” of Utah’s Canyonlands National Park, the assemblage of life-size and larger anthropomorphs would represent the norm, and only the anomalous 7-foot-tall figure called the “Great Ghost” would seem oversize. In my discussions of central High Plains rock art in previous chapters, I have provided detailed information about the size of individual engraved and painted figures at specific rock art sites. Seldom were any images disproportionately large or small; in fact, a distribution study would show that values for the height of petroglyphs and pictographs would be clustered within a fairly narrow range. There are, however, several dozen figures on the central High Plains that are clearly oversize when compared with rock art images in the region generally. One or two of the painted or pecked figures are anthropomorphs, but most are animals, usually large bears, which have been painted using black and red pigments. The best-known figure is a life-size bear located along the Purgatoire

194

L AWRENCE L. LOENDORF — THUNDER

AND

HERDS

River near Las Animas, Colorado (Figure 7.1). In the first written description of the image, appearing in 1874, the bear is identified as “Ursa Major”: his length being seven feet and his hight (sic) three feet and seven inches. His delicate ears stand up straight three inches, and his stub tail sticks out straight three inches. His cinnamon color, which penetrates the rock an eighth of an inch, is well preserved, and contrasts finely with the gray of the sandstone background.10 Besides the image’s exceptional size, the bear’s cinnamon color is also unusual and may be accounted for in at least two ways, only one of which would have been intentional. If, at the time of painting, small amounts of red pigment had been added to a black-based paint, the result would have been a dark, reddishblack color. It is also possible that the red color may have come from a natural source, since the Las Animas Leader describes a seam of red rock above the bear from which, during rainstorms, reddish water could have washed over the figure. The newspaper refers to this process as the “bleeding” of the bear and reports that when it occurred, local Indians left beads, food, and arrowheads at the site as offerings. The article also comments on the presence of a grinding stone at the base of the rock art panel on which arrowheads and tomahawks were sharpened to “actuate” the spirit of the great bear, which the weapons then absorbed. It is not clear, however, if this information came from Indian sources or from the imagination of the article’s author. Although the importance of the site in Indian spiritual life was well known to Euro-Americans living in Colorado in the late nineteenth century, the circum-

FIGURE 7.1. A stereo photograph of the Purgatoire bear taken by B. H. Gurnsey about 1889. Used with permission from the Denver Public Library.

7 — HISTORIC PERIOD INCISED

AND

PAINTED ROCK ART

195

stances surrounding the placement of the bear’s image on the rock were unclear. In some descriptions it was referred to as a freak of nature, although B. H. Gurnsey’s early stereo photograph of the image referred to it as a “natural” photograph of a bear. The figure has not fared well and, as early as the 1870s, it was damaged by bullet holes and graffiti. The vandalism prompted the Las Animas Leader to announce a plan to engrave the following warning in bold letters on an adjacent rock: CAUTION !!! Touch not a hair of that Cinnamon bear, For Nature, the goddess, has placed it there! But sure as you do, Whether Gentile or Jew. May a cinnamon bear disfigure you!11 Although the newspaper never followed through on its intention, an anonymous person did write “Do not deface the bear” on the sandstone wall above the figure. Even this bold attempt to protect the figure was unsuccessful, however, and today only remnants of it remain.12 Fortunately, the site is now on private land and is protected from further destruction. Other oversize bears in the High Plains region include a faded red, seated figure drawn in profile on the wall of a sandstone alcove. Measuring about 0.5 meter in height, the image is surrounded by vertical streaks of red paint. Another large, predominantly black bear—also drawn in profile view, probably with charcoal-based paint—has a red mouth and a circular red area behind its front leg near where its heart would be (Figure 7.2). In the far southeastern corner of Colorado, the figure of a large bear, painted with red pigment, is shown sitting upright facing the viewer. Large claws are visible

FIGURE 7.2. A large painted bear at the Red Top Ranch, Colorado. Photograph by Anne Whitfield.

196

L AWRENCE L. LOENDORF — THUNDER

AND

HERDS

on its front paws and, although they are now faded, its lower legs appear to end in claws as well. The bear occurs at a site that contains very large, nearly life-size animals that appear to represent quadrupeds. A group of these quadrupeds is found at the Hicklin Springs site, but they also occur at other sites in the Purgatoire River Valley. The figures, whose outlines have been solidly pecked, have either ovoid or rectangular bodies, straight legs, and long necks (Figure 7.3). The interior of their bodies consists of very scattered stipple-pecking. Also in southeastern Colorado, at a site on Mesa de Maya, Sally Cole has reported the presence of three life-size quadrupeds. These well-made, stipple-pecked figures, which have tails and open mouths containing teeth, may be mountain lions or perhaps wolves.13 The oversize anthropomorph category includes an unusual set of figures at the Hicklin Springs site in southeastern Colorado which appears to represent the transformation of a human into a bird.14 The four related images are in very poor condition, but enough of their original form remains to indicate that they were originally painted white (Figure 7.4). The anthropomorph, slightly more than a meter in height, has a narrow waist, square shoulders, and a round head. One leg is visible, but the other ends at a crack in the wall of the sandstone cliff. A second figure, to the right of the anthropomorph, has downward-turning, outstretched arms that resemble wings, and it is connected to a third, bird-like form with feathered wings. A fourth figure, to the right of the third, has the body of a bird and well-formed wings.

FIGURE 7.3. Oversize figures from the Hicklin Springs site in southeastern Colorado. Photograph by Lawrence Loendorf.

7 — HISTORIC PERIOD INCISED

AND

PAINTED ROCK ART

197

FIGURE 7.4. The transformation scene at Hicklin Springs. Note how body parts of the figures seem to disappear or emerge from cracks in the rock wall. Illustration by Chris Alford based on photographs by Lawrence Loendorf. An important aspect of the figures in this panel is their placement relative to the cracks in the surface of wall. As I noted, one leg of the anthropomorph appears to have disappeared into, or to be emerging from, a crack in the sandstone wall—the same crack at which the lower body of the second figure terminates—and the lower body of the third figure is contiguous with another crack higher up on the cliff face. The image of the well-made bird is painted over the cracks as though it has fully emerged from the rock face. A long, arching line connects the left side of the human figure with a rock art panel below it containing a dual-species figure with human arms and the tail of a bird. In the discussion of Purgatoire Painted rock art in Chapter 5, I referred to ethnographic sources that describe the important role played by birds in Pawnee lifeways. These accounts do not, however, contain information that would clarify the possible meaning of the figures in the preceding atypical panel, nor is there any discussion of the circumstances in which they might have been created. In the absence of such documentation, I can only speculate that the panel may have been the work of a historic Pawnee shaman painter, although it may also be attributable to another ethnic group.

A Temporal Framework for Oversize Figures At this point, our knowledge of High Plains rock art paintings is insufficient to permit the chronological placement of oversize bear figures with any confidence. More certain, however, is the fact that bears appear to be the animals most frequently represented as life-size figures. They played an important tutelary role in the spiritual life of all of the Plains Indian tribes that frequented the region during the Historic Period, so their oversize images could have been painted by any

198

L AWRENCE L. LOENDORF — THUNDER

AND

HERDS

of these groups. That said, the red-painted sitting bears might be part of the Purgatoire Painted style, but they would be atypical because of their large size. The streak marks surrounding one bear are a common element in the Purgatoire Painted style, but very similar vertical lines, both pecked and painted, are also found at Historic Period sites, so the marks do not constitute diagnostic elements. For a variety of reasons—among them the differences in the consistency and color of the bears’ red paint and that used to define Purgatoire Painted figures, the figures’ unusually large size, the fact that one of the bear images was visited by historic-era peoples bearing offerings, coupled with the stipple pecking of the large quadruped petroglyphs—I believe the oversize figures are more recent than Purgatoire Painted figures. And, because they do not occur at Apache rock art sites, I do not think the figures should be assigned to the Protohistoric Period. Based on the preceding evidence, therefore, I believe that the oversize figures in the region were created after A.D. 1725, the beginning of the Historic Period. Future researchers ought to be able to confirm this assertion by dating samples of charcoal-based pigments.

PICTURE CANYON In the Comanche National Grassland, where Picture Canyon is located, the emphasis is on looking. Thousands of visitors each year come to see the many resident species of wildlife, the dinosaur footprints, the extraordinary geological formations, and colorful sunsets. They come to escape the handiwork of man, which in this vast natural landscape is virtually nonexistent—with the exception, that is, of the rock art in Picture Canyon. By the 1930s, the peripatetic Etienne B. Renaud had discovered and been so impressed by the site’s pecked, incised, and painted figures that he published three descriptions of the figures he recorded.15 Observing that pecked figures appeared to predate the incised and painted examples, he identified what he thought were pecked images of an antelope and possibly a coyote. Renaud’s quadruped inventory also included the incised figures of a horse and rider, a buffalo, and two skunks, and one painted figure of a large black bison. Renaud noted the presence of several anthropomorphs, which included the realistic figure of a human body, the most striking petroglyph discovered this season, a woman with sign of smallpox over the body. The interpretation was made more certain after finding four similar drawings in “Battiste Good’s Winter Counts” as given by Mallery in the 10th annual report of the Bureau of American Ethnology.16 Renaud recognized that the rock art of Picture Canyon was produced by protohistoric and historic tribes, and he realized that Mallery’s invaluable ethnographic work in the nineteenth century described peoples whose activities were close enough in time to be relevant to an interpretation of that rock art.

7 — HISTORIC PERIOD INCISED

AND

PAINTED ROCK ART

199

Not long after Renaud’s visit to Picture Canyon, recreational use of the site increased, and it became a popular destination for picnics and other outings. A significant number of the visitors left their names, initials, and dates on the rock art panels, and this extensive graffiti, coupled with the rapid erosion of the sandstone surfaces, left the rock art in very poor condition. Fortunately, in 2000 a group of volunteers from the Pueblo Chapter of the Colorado Archaeological Society, led by Anne Whitfield and Mike Maselli, began a project to record the rock art remaining at Picture Canyon.17 They discovered that, in addition to the main site and an associated archaeoastronomical feature called Crack Cave, the locality actually consists of seven separate sites, a fact that was known to some residents of the area but had not been officially reported.

Anthropomorphs, Circles, and Smallpox The anthropomorph that Renaud described as a female figure with the “sign of smallpox over the body” is over a meter high and is shown in profile view. Its fleshed-out arms and legs and its round head, eyes, and mouth make it appear more realistic than most rock art representations of humans. Identification of the figure as female is based on a bulge on its right side that some researchers believe represents a breast, although this is not a convention for designating gender elsewhere on the High Plains. It is the presence of 16 or 17 circles on the figure’s body that prompted Renaud’s diagnosis of smallpox, which he based on similarities between the circles and the mnemonic icons in “Battiste Good’s Winter Count” which are known to represent smallpox. Not all researchers accept the validity of Renaud’s comparison, supporting their rejection by citing circles that are not indicative of smallpox on similar figures at other rock art sites. Bill Buckles, for example, has pointed out that a similar petroglyph in the Galisteo Basin near Santa Fe, New Mexico, is identified by Pueblo Indians as a representation of Shulawitsi, the Fire God, and that another similarly spotted figure in a painting by George Catlin represents a Mandan spirit.18 While I agree with Buckles that it is important to look for alternative interpretations of rock art—and it is clear that the endeavor is problematic—I believe that in this instance there is an attribute associated with the Picture Canyon figure, and with those in the Battiste Good drawings, that Buckles has overlooked. A small scroll design near the former figure’s mouth, and near the belly or another body part in the Good illustrations, has been identified as an indicator of pain. This pain symbol is not found near the figures of either the Pueblo Fire God or the Mandan spirit, an important omission indicating they are not as similar to the Picture Canyon figure as Buckles suggests. More importantly, the combination of spots or circles and the pain symbol strengthens the argument that the Picture Canyon figure represents an individual with smallpox. The presence of a very similar scroll

200

L AWRENCE L. LOENDORF — THUNDER

AND

HERDS

symbol in an Aztec drawing of smallpox also suggests that the symbol was accepted as an indication of pain across a large part of North America (Figure 7.5). The presence at Picture Canyon of a 60-centimeter-tall figure in a panel adjacent to the spotted anthropomorph, also discussed by Renaud, may be helpful in determining whether or not circles are synonymous with smallpox.19 This anthropomorph is situated within a group of 25 to 35 circular depressions that closely resemble the pecked depressions, or cupules, seen in other rock art contexts (Figure 7.6). Varying in extent from walnut-size to baseball-size, cupules occur in rocks and boulders all over the world. Speculation about what they may be ranges from the practical—“probably used to grind plants or seeds”—to the imaginative—“possibly represent stars in a map of the cosmos.” While either of these two conjectures might account for the form and purpose of some cupules, their ambiguity makes them an intriguing subject for research. The examples at Picture Canyon are particularly unusual because they occur on a vertical sandstone surface, in contrast to most High Plains cupules, which are found on flat surfaces.

FIGURE 7.5. On the left is the anthropomorph at Picture Canyon that may be suffering from smallpox. On the right is a composite illustration showing the Battiste Good drawings of smallpox and the Aztec pain symbol. Redrawn by Elaine Nimmo from Mallery 1972:308, 313; and Historia De Las Cosas de Nueva Espana, Volume 4, Book 12, Lam. cliii, plate 114. Peabody Museum of Archaeology and Ethnology, Harvard University.

7 — HISTORIC PERIOD INCISED

AND

PAINTED ROCK ART

201

FIGURE 7.6. Anthropomorphic figure at Picture Canyon surrounded by cupules. Illustration by Anne Whitfield.

The encircled anthropomorph has one cupule where its head would normally be and another between its legs which may represent a vulva. Two smaller cupules on its chest might represent breasts, but the presence of chalking and graffiti on the panel makes it difficult to tell whether they are original features of the petroglyph or later Euro-American additions. The figure is conventional in other respects, with a rectangular body and upraised arms terminating in fingers. Its

202

L AWRENCE L. LOENDORF — THUNDER

AND

HERDS

straight legs end in L-shaped feet that point away from the midline of the body. Many of the cupules surrounding the anthropomorph have attached incised lines that extend downward like the strings on balloons. Some cupules are bisected by these lines, and multiple lines are connected to at least one cupule. What the lines represent is unknown, but they are not usually associated with cupules at other sites in the region. The proximity of the figure within the cupule panel to the spotted anthropomorph suggests to me that the two anthropomorphs may be related. Cupules are sometimes said to be features created by shamans or other individuals during the grinding of geologic surfaces to obtain powdered rock for medicinal purposes. It is tempting to speculate that the sandy residue from the cupules at Picture Canyon may have been used as medicine for treating the diarrhea and fevers associated with smallpox.20 An alternative explanation offered by local ranchers—and prevalent at other localities throughout the American West—identifies the sandstone surface as a mother rock to which pregnant women would come and make offerings. Such a view is compatible with the possibility that pregnant woman may also have used the residue for medicinal reasons.21

Another Anthropomorphic Figure at Picture Canyon Controversy surrounds another significant anthropomorph at Picture Canyon. This figure has a rectangular body, short neck, and round head that includes eyes and a mouth (Figure 7.7). Its legs descend from a bow-shaped crotch and terminate in L-shaped feet.22 Attached to its upraised arms are spears or arrows with leaf-shaped points oriented toward the ground. A prominent hourglass design across the figure’s chest is connected to its right shoulder, as though it is part of a decorative sash or figured shirt. The lines used to form the figure’s shoulders are rounded and then continue downward in straight lines. In this respect, as some researchers have claimed, the figure is similar to the V-shouldered or Vnecked anthropomorphs that are prevalent in eastern Montana, northeastern Wyoming, and western South Dakota. Examination of early photographs of this anthropomorph makes clear, however, that it differs in significant ways from the classic V-necked figures of the northern Plains.23 For one thing, its neck is made with double vertical lines, whereas the necks of classic V-necked figures consist of only one straight line. Two other important differences include shoulder lines that connect to the sides of the neck instead of meeting in a V-shape at the throat, and the fact that the right shoulder line continues past the throat before connecting with the hourglass design on the figure’s chest. While the distinctive physical attributes of the Picture Canyon anthropomorph are sufficient to distinguish it from Plains V-necked figures, in order to determine what the figure actually represents we now have to refer to ethnographic sources for information about what the two downward-pointing spears

7 — HISTORIC PERIOD INCISED

AND

PAINTED ROCK ART

203

FIGURE 7.7. A drawing of the twolance figure at Picture Canyon based on a photograph in Thomas (1929:292). The figure is clearly not a classic Plains V-necked anthropomorph. Drawing by Elaine Nimmo.

might mean. From those sources we learn that it was customary among Plains Indian warrior societies, where acts of bravery were recognized as part of intersocietal competition, for a warrior to define a battle position by staking his spears in the ground. Among some tribes, once the spears were in position, there was no retreat unless a fellow-warrior removed them. Warrior societies often acted as police during buffalo hunts, a role fulfilled among the Pawnee—who may have connections to Picture Canyon—by a Two Lance Society and a Red Lance Society. Warriors from these societies staked their lances in front of their tipis as soon as a hunting camp was established, making their weapons available for other hunters to use in chasing buffalo. Alfred B. Thomas, noted for his historical research on Spanish–Indian relations in New Mexico, was the first researcher to point out the resemblance between the posture of the Picture Canyon figure and that of a warrior holding lances.24 Thomas based his identification on small character glyphs representing Comanche warriors that appeared on a 1786 Spanish tally sheet drawn up by New Mexico Governor Anza. The tally sheet documented the assistance given by Comanche warriors to Spanish forces fighting the Apache in the Sandia Mountains of central New Mexico. There is no reason to suspect that the Picture Canyon figure represents a Comanche warrior, however, since the practice of staking lances was widespread.

Horse and Rider Figures at Picture Canyon Several of the Picture Canyon figures depict riders on horseback. In one figure, an incompletely incised outline is partially filled with black, probably charcoalbased, pigment. This figure appears to be in the center of a tipi camp, which is

204

L AWRENCE L. LOENDORF — THUNDER

AND

HERDS

FIGURE 7.8. A human figure standing on the back of a horse at the Picture Canyon site. Illustration by Anne Whitfield.

indicated by simple, inverted V-shapes. The human figure appears to be standing upright on the back of the horse, which has a relatively long body and a long, arching neck (Figure 7.8). This posture is common in very early depictions of Plains Indian horse-and-rider figures and has been explained as a lack of mastery of perspective by the artists, who were unable to create the image of a rider with a leg on both sides of a horse. Unfortunately, much of the detail of the rider in the Picture Canyon image has either been destroyed by erosion or is obscured by smudging, making it difficult to compare the figure with a similar horse-and-rider figure at the Tolar site in southwestern Wyoming.25 The legless torsos of both riders appear pointed and rest on the backs of their respective horses; their arms seem to be upraised and end in hands that hold objects. The Tolar figure holds a shield and a lance, while the Picture Canyon rider may have a lance close to its left hand, but there is no associated shield. The horses at the two sites are also similar, and both have long, thin necks. A comparison with drawings made by the Comanche has prompted researchers to assign authorship of the figures at the Tolar site to that group. Even though the smudged drawing at Picture Canyon lacks clarity, I would not be surprised if it, too, were the work of Comanche artists.

7 — HISTORIC PERIOD INCISED

AND

PAINTED ROCK ART

205

Tool Grooves and Quadrupeds Hundreds of parallel, incised depressions commonly identified as tool grooves occur among the many figures at Picture Canyon. Used to sharpen bone awls and other tools, most grooves are clustered in random groupings, but there are a few instances in which they occur in patterned rows. One example consists of a long central line crossed at right angles by parallel grooves whose length seems to depend on the space available on the rock surface.26 This design resembles the ribstones pecked on rocks along the Canadian border to enhance a replica of an animal believed to be contained within the rock itself (Figure 7.9). In the north, these stones were regarded as fetishes containing the spirits of bison, and they were often regarded as prayer stations by Siouan- and Algonkian-speaking tribes. A somewhat similar row of parallel lines forming a rib-like pattern occurs at the Star site in Kansas.27 A variety of quadrupeds are represented that, in addition to the horses, include bears and bison drawn with black pigment. One especially large black bison measures 2.7 meters from nose to tail, but it is now in such poor condition that it is very difficult to see.28 Renaud doubts the authenticity of this large quadruped, but the presence of other large black figures in the region suggests it is probably Indian-made. The head and open mouth of a partially drawn bear is depicted as though it is emerging from the rock in what researcher James D.

FIGURE 7.9. A ribstone at Picture Canyon. Illustration by Anne Whitfield.

206

L AWRENCE L. LOENDORF — THUNDER

AND

HERDS

Keyser has termed the “Bear Coming Out” motif.29 According to Keyser, such figures represent bears emerging from their dens. Based on the relatively large number of images of horses at Picture Canyon, the majority of the petroglyphs appear to date to the Protohistoric and Historic Periods.30 Charcoal-painted figures in open-air sites on the Plains are also assumed to have been produced in later time periods. The tool grooves and patterned, abraded lines, as well as the human figures with rectangular bodies, may belong to the Late Prehistoric Period, but they may have been created more recently as well. Very few of the Picture Canyon figures are grouped in the kind of dynamic narrative sequences typical of the Biographic tradition prevalent on the Plains by the end of the Late Prehistoric Period and discussed later in this chapter. Although figures of horses and riders in one panel may be pursuing a bison—a theme at other Historic sites in the region—and in another panel, a stick-like charcoal figure is attached by a rope to a horse, perhaps indicating that he has captured it, the majority of the figures are static and not engaged in identifiable actions.

Crack Cave Crack Cave is the most well-known feature at Picture Canyon and certainly the most controversial because of claims that have been made about the meaning of the images on its walls. The site is located along the eastern side of Picture Canyon, within the boundaries of an abandoned Historic-era homestead. There are four rock art panels at the site, two in rockshelters and two in open-air settings. The “crack” that gives the site its name is actually a crevice, wider at the bottom but rapidly narrowing to an opening about the width of a person, which continues upward for 6 meters or more and provides access to an interior area that is approximately 6.5 meters deep. The surface of the site contains Historicera trash and a relatively dense scatter of prehistoric chipped stone tools. No excavation has been undertaken to investigate whether the site has significant time-depth or to find any buried features that would support or challenge claims for the origin of the rock imagery. Like the Sun Temple site, also in southeastern Colorado, Crack Cave is believed by amateur epigraphers to contain inscriptions in a written Celtic language termed “Ogam” or “Ogham.” Supposedly recorded on five or six panels (depending on who is doing the translation) during journeys to Colorado by ancient Celtic voyagers sometime between A.D. 350 and 900, the “text” has been translated as “SUN STRIKES HERE ON THE DAY OF BEL,” announcing that the cave is illuminated at dawn during the vernal and autumnal equinoxes.31 According to Ogam researchers, the statement was translated before they knew that a shaft of sunlight indeed penetrates the cave and illuminates the markings twice each year. As for the images themselves, a petroglyph of a rayed arc is interpreted by some researchers as a set of Ogam characters, but the pattern could also repre-

7 — HISTORIC PERIOD INCISED

AND

PAINTED ROCK ART

207

sent the sun. Other figures include parallel sets of vertical incised lines, a box-like form with extended sides, a group of lines that intersect at odd angles, and a series of fan-shaped incised lines. The lines have been altered, either by human abrasion or animal rubbing, and are in poor condition. Partly in response to destructive agents, the cave is now protected by an iron gate. The controversy at Crack Cave is not about the suggested connection between the rayed petroglyph and the sun’s penetration of the cave at sunrise on each equinox. Rather, the debate concerns whether the incised lines represent an Ogam inscription that foretells such events. It is well known that the solstices and equinoxes were important to Plains Indian tribes, who used geologic features and rock art as calendars. Abundant evidence of prehistoric astronomical reckoning comes from features such as the council circles in central Kansas, the medicine wheel in Wyoming’s Bighorn Mountains, the orientation of the Mandan Okipa ceremonial lodge, the painted black buffalo in Teton Jackson cave, Montana, the orientation of the mounds at Spiro, Oklahoma, and the C. C. Witt site in central Kansas.32 The list of astronomical features is extensive, but these examples are sufficient to demonstrate that ceremonies marking the solstices and equinoxes were well integrated into Plains Indian lifeways many hundreds of years before the arrival of Euro-Americans in North America. Considering this tradition, one has to wonder why the search for alternative explanations of features used by First Americans for astronomical computation has gained such momentum. I think one reason is related to cultural bias. People simply have difficulty accepting that American Indian cultures were sophisticated enough to appreciate what the solstices represent. As evidence I cite the public and scholarly reaction to the discovery of a petroglyph calendar at the Fajada Butte site in Chaco Canyon. The announcement prompted worldwide television coverage and a debate among Euro-American researchers, some of whom claimed it might be genuine and others who thought the solstice spiral was a naturally occurring phenomenon rather than an intentionally placed design.33 Interestingly, when Hopi and other Pueblo peoples were shown the petroglyph and asked to assess its significance, they responded offhandedly that these sorts of petroglyphs occur all across their homelands.34 I strongly suspect that if a group of Pawnee or Wichita elders were taken to Crack Cave and asked about the significance of the petroglyph that is illuminated by the sun on the equinoxes, their reaction would be similar. Occam’s Razor is a time-tested rule in research that can be summarized as favoring the least complicated of two or more explanations for a phenomenon, other things being equal. In the case of the petroglyph and incised marks at Crack Cave, their explanation does not require a complex hypothesis involving inscriptions made by ancient Celtic visitors, because a well-documented tradition of aboriginal astronomical observation already exists. A comparison of the Crack Cave petroglyph to artifacts and features at the C. C. Witt site illustrates this point.

208

L AWRENCE L. LOENDORF — THUNDER

AND

HERDS

The site, identified by excavator Patricia O’Brien as an ancestral Skiri Pawnee earth lodge in use at about A.D. 1300, contained the remains of animal body parts important in Pawnee ceremonies: a redheaded woodpecker, a blue jay, a rare long-eared owl, four bobwhite quail, a woodpecker of unknown species, the talon of a bald eagle, five gar fish skulls, a snapping turtle carapace, a painted box turtle carapace, and four unmodified mussel shells.35 These artifacts were almost certainly once part of a prehistoric medicine bundle and, recognizing this association, O’Brien turned to astronomer and ethnohistorian Von del Chamberlain’s detailed study of Pawnee cosmology.36 There she discovered that the long axis of the Witt site earth lodge was oriented so that on the vernal equinox, “dawn sunlight would shine in through the east entrance and pass over the central fireplace to strike the altar.”37 Similar Pawnee ceremonial associations are present at Crack Cave, which is thought to be contemporaneous with the Witt site. For instance, the cave entrance—the equivalent of the ceremonial lodge door—is open to the equinox sun, which shines on an abstract petroglyph (Figure 7.10). It is widely accepted that Pawnee and other Caddoan-speaking groups made hunting expeditions into Colorado, and it is perfectly reasonable that, in such a context, a priest would have inscribed the Crack Cave petroglyph as a dual-purpose symbol. It may have represented a ceremonial altar, and it would also have reminded the group when they needed to perform important rituals to renew their world. The future discovery, although unlikely at this point, given the destructive human presence in the cave over time, of ceremonial bird bones, gar skulls, or turtle carapaces would definitively place prehistoric Pawnee travelers, rather than Celtic voyagers, in Crack Cave.

FIGURE 7.10. The abstract pattern in Crack Cave, southeastern Colorado. Illustration by Anne Whitfield.

BIOGRAPHIC ROCK ART Although variability in the subject matter and types of rock art figures characteristic of the Historic Period on the central High Plains makes them difficult to categorize, the majority are encompassed by what archaeologist James Keyser has

7 — HISTORIC PERIOD INCISED

AND

PAINTED ROCK ART

209

identified as the Biographic tradition.38 Biographic rock art includes recognizable narratives with “story lines” that are often related to the widespread Plains Indian practices of horse-raiding and warfare. Figures of horses, guns, and a variety of other objects are displayed in ways that help the viewer understand how these objects are related to an event, such as the capture of an enemy gun (Figure 7.11). The event has usually taken place in the past, but in some instances individuals would visit a rock art site and add an image as an indication that the event would be repeated at some future time. The same conventions for representing events and sequences of actions are found in the Biographic rock art of all regions of the North American Plains.39 Many figures tell the story in a form of Plains Indian shorthand: for example, horse tracks and human footprints indicate the direction that a group entered and then left a village, and a simple, inverted V representing one tipi was understood to indicate the presence of others.40 The primary difference between Biographic rock art on the central High Plains and elsewhere relates to the techniques used to produce the figures. On the northern Plains, incised and charcoal lines predominate, while on the central High Plains, in addition to these techniques, figures may be pecked or abraded, painted in red, or painted in red and black. Pecked human figures occur, either on horseback or as freestanding figures, but pecked images of horses and other historic objects are unusual. Pecked images contradict the expectations of many Plains rock art researchers, in whose model of technical progression incised petroglyphs occur more recently than pecked figures. Several instances in which empirical evidence has challenged this sequence include the incised figures that underlie pecked petroglyphs of

FIGURE 7.11. Typical Biographic image of horse and rider. The figure is at the Cairn site on the Piñon Canyon Hogback. Tracing by Linda Olson.

210

L AWRENCE L. LOENDORF — THUNDER

AND

HERDS

Archaic age at the Ancient Hogback site and the images of incised nets that occur underneath quadruped figures at several Developmental Period sites. Nevertheless, a general rule applying to the prehistoric traditions on the central High Plains is that incised figures are more recent, which is why the pecked figures of horses at the Red Guns site on the lower Purgatoire River are so unusual. Among the many images at the Red Guns site are 27 painted red guns, probably representing captured weapons, that both over- and underlie images of horses, riders, and anthropomorphs unassociated with horses. In one panel, a pecked horse with what appear to be multiple riders is superimposed over several red guns. At another location, the outline of a pecked and abraded human head, with large round eyes and an abraded, cupule-like mouth, occurs underneath images of red guns (Figure 7.12). Near the head, the upper torso of a human figure with a bucket-shaped, horned head overlies the guns. A second anthropomorph with a rectangular body and similar horned, bucket-shaped head is also superimposed over images of guns. The latter figure has outstretched arms whose upturned hands obscure portions of the guns. Historical sources provide chronological information that helps determine the sequence in which the preceding images were created. By A.D. 1700 tribal

FIGURE 7.12. Two illustrations of a set of red-painted guns at the Red Guns site. The larger image of the panel is made from an enhanced photograph. The inset line drawing is presented for clarification. The guns are superimposed over a round head but underlie the bucket-shaped head of the figure on the right. The guns are approximately 60 centimeters long. Inset drawing by Elaine Nimmo from a photograph by Lawrence Loendorf.

7 — HISTORIC PERIOD INCISED

AND

PAINTED ROCK ART

211

groups on the central High Plains, especially the Apache, possessed horses, and many groups had obtained and were using guns by about A.D. 1750. Based on the placement of figures either under or on top of motifs of these temporal markers, it can be said that, for instance, the pecked and abraded human head is likely to predate A.D. 1750, whereas the pecked horse with multiple riders and the anthropomorphs with bucket-shaped heads postdate A.D. 1750. The sequence becomes even more interesting when the figures on a panel immediately below the guns are examined. The primary image is a pecked and abraded horse measuring 92 centimeters from head to tail, on whose back a human figure with a round head and rectangular body has been set (Figure 7.13). This legless human has distinctive weeping eyes and an abraded cupule-like mouth. Two additional humans with rectangular bodies and round heads, one eyeless and the other with weeping eyes, also occur on the panel. The left motif is superimposed over the figure on the right, which is itself superimposed on another very round head having large ears, circular eyes, and a cupule-like mouth that is otherwise very similar to the head that overlies it, over which guns have been painted. Attributes shared by the large horse and rider, the solitary round head, and the anthropomorphs with round heads, cupule-like mouths, and weeping eyes indicate that all of these images are contemporaneous. The presence in the group of a horse, which was only present in the region after about A.D. 1700, and the fact that guns were only available after A.D. 1750, make it quite likely the group of figures was made between A.D. 1700 and 1750. Elsewhere on the Plains, large round heads and weeping eyes are associated with the Comanche.41 Although typical of this class of motifs in most respects, the large round head with ears, nose, and a mouth at the Paradise site in western Kansas is unusual because the body lacks a base line.42 The deeply incised and FIGURE 7.13. A drawing of the pecked horse and rider at the Red Guns site. The horse measures 92 centimeters from head to tail. This image is immediately below the superimposed head and guns shown in Figure 7.13. Drawing by Elaine Nimmo from a photograph by Lawrence Loendorf.

212

L AWRENCE L. LOENDORF — THUNDER

AND

HERDS

abraded figures at the Red Guns site, which date to the time when the Comanche were present in the region, also lack base lines, which makes me suspect that they, too, were made by the Comanche. The bucket-shaped heads are more difficult to link to known ethnic groups. They are not commonly found in the region, nor do they appear in sites in western Kansas or elsewhere that contain incised anthropomorphs. One possibility is that they represent the work of Plains Apache, who made figures with pecked outlines like those described in Chapter 6. In spite of this unresolved question, these pecked figures are important because they have stylistic properties that in other settings would almost certainly be classified as older, yet they occur very late on the central High Plains. Other figures at the Red Guns site, including several situated in the vicinity of those I have already discussed, consist of lightly incised horses and riders. One human standing on a horse wears a long flowing headdress, while to the rear of the horse there is a solitary horned head (Figure 7.14). The horse and rider, along with a half-dozen other figures at this site, the finely incised figures at Apache Creek, New Mexico, and on the Hogback at Piñon Canyon, closely resemble classic examples of the Biographic style on the northern Plains.

FIGURE 7.14. A small incised horse with standing rider at the Red Guns site. Some researchers believe the winged figure behind the horse may be a cradleboard. Drawing by Elaine Nimmo from a photograph by Lawrence Loendorf.

TECHNIQUES AND FORMS : PECKING, ABRADING, OVERSIZE , AND UNIQUE FIGURES I suspect that images of pecked horses were made by earlier immigrants to the central High Plains, like the Comanche or Ute, who brought this rock art tradition to the region. The appearance of finely incised horses, however, probably represents the arrival of tribes such as the Cheyenne and Kiowa, who moved to the central High Plains later in time. The belief that pecked horses occur earlier on the central High Plains is supported by the relatively large number of images of armored hors-

7 — HISTORIC PERIOD INCISED

AND

PAINTED ROCK ART

213

es in the region. Archaeologist Mark Mitchell has described a number of Purgatoire River region horse motifs that appear to be wearing Spanish-style armor.43 One is an oversize figure formed by deeply grooved lines that appear to have been formed by both pecking and abrading. Mitchell believes the figures of armored horses, dating between A.D. 1700 and 1750, represent the work of the Comanche, a view that is compatible with my suggestions about the ethnic identity of the makers of pecked horses during the same time period on the central High Plains. While I think it is possible that the large figures of horses and riders at the Red Guns site are close to one another in age, to say on the basis of similar size that all oversize figures are contemporaneous oversimplifies the complex problem of chronological placement. The one defensible generalization is that oversize figures clearly date to the Historic Period. The most impressive oversize figures occur in the previously described transformational panel at the Hicklin Springs site, where the figure of a painted anthropomorph appears to change into a bird. That figure is positioned to indicate that one of its legs remains within a crack in the sandstone surface and demonstrates the importance of the relationship between an image and the surface on which it has been placed. The theme of transformation from one species to another is also significant because of its rarity in High Plains rock art, although tales about humans who change into birds occur among various Plains groups, particularly the Pawnee.44 Although the panel is so eroded that it is difficult to see many of its components, its poor condition may be an indication that it dates to the early part of the Historic Period. Much more work needs to be done, however, before the placement of these large paintings in the regional chronology can be definitively established. As a group, images of large bears are the best-known figures in the oversize category, but the inventory includes large bison and antelope as well. At Picture Canyon, the figure of a bison that measures 2.7 meters from head to tail and 1.6 meters in height is very difficult to see, but researchers Anne Whitfield and Michael Maselli located old photographs in which the image was sufficiently distinct to permit them to make a drawing of it.45 Depicted in profile view, the bison has a prominent hump, two well-formed horns, and an elongated snout. The eroded legs are indistinct, but the tapering body’s erect tail is visible. In terms of technique, the bison is similar to the image of the large cinnamon-colored bear in the lower Purgatoire River drainage near Las Animas, Colorado, described earlier in this chapter. The presence at Picture Canyon of an anthropomorph with markings indicating that it may have contracted smallpox is, to my knowledge, the only known instance of such a rock art figure in North America. Earlier in this chapter I discussed the alternative interpretation of the figure’s meaning advanced by archaeologist Bill Buckles, and I countered his interpretation by drawing attention to a small scroll design near the figure’s mouth that occurs as a symbol of pain in many different cultures. My argument is not conclusive, however, because the scroll symbol is

214

L AWRENCE L. LOENDORF — THUNDER

AND

HERDS

not exclusively associated with smallpox. It occurs, for example, on Battiste Good’s winter count of 1826–1827 associated with a Dakota war party that encountered and ate the meat remaining on an old buffalo carcass.46 All members of the group subsequently suffered severe intestinal distress and eventually died.

RIBSTONES AND TOOL GROOVES The long incised line at Picture Canyon that is bisected by shorter lines is another intriguing figure. I noted in the site description that I believe this figure, which is associated with a series of tool grooves, may represent a ribstone. Distributed across the Great Plains, the most spectacular ribstones—and the ones with the best ethnographic information about their use—occur in southern Alberta and Saskatchewan and in northern Montana and North Dakota.47 Manufactured from glacial erratic boulders, ribstones are usually made of quartzite. The largest examples occur on stationary boulders that measure nearly 2 meters long by 1 meter high, although many smaller, portable examples have also been found. Ribstones may vary in their details, but all consist of a long, vertical line or groove along the length of a boulder that is crossed by shorter grooves, creating a figure that represents the backbone and ribs of a buffalo. The grooves have been pecked and abraded into the boulder surface to a depth of 1 or 2 centimeters, and a series of cupule-like holes have been placed in between the lines. The inclusion of pecked eyes, ears, a mouth, and horns suggests a living buffalo, and the presence of buffalo hoofprints on a number of boulders creates the impression of movement. Plains groups like the Cree believed that ribstones embodied the spirit of a bison, which they honored by leaving offerings and saying prayers at sites where the stones occur. The holes between the rib lines represented wounds and were said to protect the interior spirit by allowing an arrow or bullet to pass though the stone.48 In historic times, the offerings left at ribstone sites include colored cloth or peeled and painted willow sticks. Prior to the availability of cloth, tanned animal hides were left as offerings, and occasionally the Cree and other Plains groups left a deeply personal offering—a fragment of their own skin. Among Algonkian-speaking groups like the Blackfeet, small buffalo effigies, called iniskims, were placed in medicine bundles together with other holy objects. These miniature ribstones were often made from fragments of fossil baculites or ammonites, many of which resemble a four-legged animal.49 In some instances, iniskims were modified to emphasize their resemblance to living buffalo by the addition of lines representing a backbone and ribs. Prior to setting out on a hunt, owners of iniskim bundles would use them to call upon buffalo spirits and encourage them to reveal their location. It is likely that the ribstones found across the central High Plains were used as prayer stations by the Pawnee or other groups of buffalo hunters. The petro-

7 — HISTORIC PERIOD INCISED

AND

PAINTED ROCK ART

215

glyph inventory at the Star site in western Kansas, which is typical of many ribstone sites, includes a figure consisting of a series of short, parallel lines that connect to a V-shaped hump or ridge in the panel surface.50 The element for which the site was named is represented by a number of incised equilateral crosses that represent stars. Other figures include a set of turkey tracks and two human figures with their arms upraised, one of which holds a star in one hand and appears to represent a shaman. Kansas archaeologist Donald J. Blakeslee, who has written about the Star petroglyph site and its surroundings, has noted that its location is near the impact crater of the Brenham meteorite. The site contains a seam of hematite that was used for a variety of purposes by prehistoric and historic Indian groups and played a special role at shrines visited by buffalo hunters. Further evidence of the linkage between ribstone features and buffalo is provided by the placement of the Star site next to a Pawnee trail along which buffalo hunters would have passed on their hunting forays.51 At the Picnic site, south of Las Animas, Colorado, a variation of the classic ribstone feature is oriented vertically along a detached slab of sandstone. A broken portion of the same slab resembles the forequarters and head of a buffalo, a likeness that, coupled with the orientation of the ribstone, creates the impression that a buffalo is partially contained within the rock outcrop (Figure 7.15). The widespread belief on the Plains that buffalo had subterranean homes either in the earth, within caves, or inside rock formations is reinforced at the Picnic site, where the feature has been constructed to indicate that the animal is emerging from the sandstone cliff surface.52 Many other ribstone features do not as thoroughly replicate the form of a buffalo, but simply use a natural feature to convey the impression of the buffalo’s backbone and recognizable hump.

FIGURE 7.15. A ribstone on a detached fragment of the canyon wall at the Picnic site, southeastern Colorado. Note how the detached rock wall has the appearance of a bison head and back. Illustration by Anne Whitfield from a photograph by Lawrence Loendorf.

216

L AWRENCE L. LOENDORF — THUNDER

AND

HERDS

Earlier in this chapter I discussed features at Picture Canyon that were similar to ribstones, and in Chapter 4 I focused on tool grooves that occur in a domestic setting at the Red-tail rockshelter and noted the context of their probable creation and use. Now I want to expand the discussion to include those tool grooves that are often found in association with ribstones across the West in North America. In such contexts, they consist of abraded or incised lines, sometimes oriented in parallel rows and in other instances forming cross-hatched patterns. The grooves, which are usually less than 4 centimeters deep, range in length between 15 and 30 centimeters and vary in width from 1 to 5 centimeters. One characteristic that almost all tool grooves share is their placement on sandstone outcrops, although in rare instances they may occur on limestone, basalt, or other rock substrates. Most archaeologists agree that tool grooves have resulted from the manufacture of bone, antler, and wooden awls, knife and scraper handles, and bone projectile points, but as recently as the 1970s they were considered mysterious, perplexing, and difficult to interpret. Archaeology has Kenneth Feyhl to thank for his recognition of a problem in need of a solution and his development of a series of problem-solving steps that included experimental research. Feyhl, a retired petroleum geologist living in Billings, Montana, began his research by reviewing the geological literature and inquiring of other geologists working in remote places around the world whether they had seen similar marks on the rocks they studied.53 He learned from his correspondents that not only are these linear grooves found across much of the western United States, but that very similar features occur in such diverse places as Brazil, the Solomon Islands, Egypt, and Ghana. Especially interesting was the fact that some sites in Brazil were still used by native peoples for shaping and sharpening stone axes. In the journals of Peter Fidler, the chief surveyor and mapmaker for the Hudson’s Bay Company in the 1790s, Feyhl found a reference to the use of stone surfaces in tool manufacture by Kutenai Indians of western Montana and southwestern Alberta. Fidler reported that the Kutenai made wedges and chisels by rubbing the antlers of red deer back and forth across a rock surface.54 Fidler’s red deer, also called elk or wapiti, is the cervid whose massive antlers were used for making a variety of tools, among them digging sticks and scraper and knife handles. Feyhl took the additional step of making three experimental tools: a scraper handle from an elk antler, and two bone perforators, one from a deer cannon bone and the other from a deer antler. The perforators were patterned after archaeological examples found at sites in Montana, and the scraper handle was similar to ones used by Plains Indian groups. After a relatively short time, Feyhl was able to replicate historic and archaeologically recovered artifacts and, of particular interest to rock art researchers, in the process he created exact replicas of the tool grooves found on rock surfaces across North America.

7 — HISTORIC PERIOD INCISED

AND

PAINTED ROCK ART

217

Feyhl’s experiments also allowed him to answer the vexing question of why tool grooves often occur in such profusion at a site. He learned that because a groove loses its effectiveness when it reaches a particular depth, the toolmaker abandons the groove once that point is reached and begins to work on a fresh surface. The prevalent pattern of grooves that cross-cut one another was also explained: grooves are more effective if kept free of sand and grit, and cross-cut placement keeps these by-products from accumulating, particularly when a groove is placed on a slanting or vertical surface. Building on Feyhl’s research into the mechanics of tool groove production, Linea Sundstrom has explored Plains ethnography to answer the question of why they occur where they do.55 She links grooves to the buffalo hoofprints and vulva forms that are found in combination at many sites across the northern Plains, pointing out that hoofprints and vulvae are symbolic of women, the givers of life, who made many of the tool grooves during the process of sharpening bone awls. Sundstrom’s careful use of ethnography has led her to suggest that the grooves resulted from two activities among young women—puberty rituals and the preparation of tools needed by all good quill workers. Sundstrom believes that the back-and-forth motion involved in tool sharpening may be related to other repetitive, rhythmic motions used by individuals to enter a trance state. She envisions a scenario in which girls on the threshold of becoming adults would go to secluded, confined places where they made tool grooves as part of their first menstruation ritual. Such behavior would account for the presence of some grooves on the ceilings of small caves rather than on more readily accessible sandstone surfaces. Sundstrom also suggests that during menstrual rituals, young women had visions in which new quillwork and beadwork designs were revealed. Citing Lakota and Dakota ethnography, she states that these visions might be related to Double Woman, an important supernatural being who was especially skilled at quillwork. Sundstrom’s research indicates that many tool grooves on the northern Plains result from the industrious work of women as they made decorated hides. I have pointed out elsewhere in this book that tool grooves were the byproduct of both male and female toolmaking and maintenance activities. I suspect that those associated with ribstones and found at hunting-related shrines like the Star site were created by males as they prepared to hunt buffalo by sharpening their wooden, horn, bone, stone, and metal tools. The manufacture of horn bows was also a male activity, one that required considerable reduction of the horn’s outer sheath to get access to its pliable inner core. Many kinds of wooden tools such as digging sticks required shaping and sharpening throughout their use life. Since the majority of tool grooves occurring on sandstone surfaces resulted from these kinds of toolmaking activities, it would be interesting

218

L AWRENCE L. LOENDORF — THUNDER

AND

HERDS

to extend the research begun by Feyhl to determine whether the processing of different materials might create particularly distinctive properties that could be observed at either the micro- or macro-level. These studies have the potential to further illuminate the gender of the toolmaker and the activities involved.

NOTES 1 Price 2006:5. 2 Jansen et al. 2002. 3 Secoy 1953 and Ewers 1955 remain excellent sources of information about the role of the

horse in Plains Indian lifeways.

4 Kavanagh 1996:68; 2001. 5 Callaway et al. 1986:352. 6 Shimkin 1986:517–18. 7 Rollings 1989:32. 8 Hamalainen 1998. 9 Keyser (2004) discusses the variety of historic rock art images across the Great Plains. 10 Las Animas Leader, August 14, 1874, p. 1. 11 Las Animas Leader, August 14, 1874, p. 1. 12 See Cassells (1997:260) for a 1929 photograph of the bear, and Buckles (1989:150) for a pho-

tograph of the bear taken in 1989. A comparison of the two photographs reveals that much of the destruction took place in the first two decades of the twentieth century. 13 Cole 1985:12, Fig. 9, p. 106. 14 McGlone et al. 1994:88. 15 Renaud 1931a:68; 1931b; 1936:Pl. 13. 16 Renaud 1931a:68. 17 Mitchell 2002:1. 18 Buckles 1989:121–22. 19 Mitchell 2002:39–41. 20 See the paper by Kevin Callahan presented at the 2000 Annual Meeting of the Society for American Archaeology. 21 See Young (1988:177) for a discussion of the mother rock near Zuni, New Mexico. 22 Mitchell 2002:30–31. 23 See Thomas (1929:292) for an early photograph of the figure. 24 Thomas 1929. 25 Loendorf and Olson 2003. 26 Mitchell 2002:29. 27 Blakeslee 2003. 28 Mitchell 2002:36. 29 Keyser 2004. 30 Mitchell 2002:56. 31 McGlone et al. 1994:96; McGlone et al. 1999:50–53. 32 Wedel 1967; Eddy 1974; Loendorf and Conner 2004; O’Brien 1986. 33 Zeilik 1985. 34 Stoffel et al. 1994. 35 O’Brien 1986. 36 Chamberlain 1982. 37 O’Brien 1986:941.

7 — HISTORIC PERIOD INCISED

AND

PAINTED ROCK ART

219

38 Keyser 1987, 2004; Keyser and Klassen 2001. 39 Keyser and Klassen 2001; Klassen 2005. 40 Keyser and Mitchell 2000. 41 Loendorf and Olson 2003. 42 Wells 1996. 43 Mitchell 2004b. 44 Harrod 2000:73. 45 Mitchell 2002:Panel 17. 46 Mallery 1972:589. 47 Wormington and Forbis 1965; Hoy 1969. 48 Wormington and Forbis 1965. 49 Wormington and Forbis 1965. Blackfeet iniskims are illustrated in Scriver (1990:178–79). 50 Blakeslee 2003. 51 Blakeslee et al. 1986; Blakeslee 2003; Blasing 1993. 52 See Chapter 4, note 25. 53 Feyhl 1980. 54 MacGregor 1966:78. 55 Sundstrom 2002, 2004.

8 THROUGH A G L A S S, D A R K L Y

n the 1985 movie The Left Handed Gun, the herd boss reads the often-quoted verse from I Corinthians 13—“for now we see through a glass, darkly”—to Billy the Kid. “The Kid” (played by Paul Newman) responds, “Does that mean like looking through a whiskey bottle?” The recognition of how deficient human powers of perception often are— and how relevant this reality is to a discussion of rock art research—occurred to me when I was thinking about a title for this chapter. I was also guided in my choice by the work of the eminent archaeologist Waldo Wedel, who used the same title for the final chapter of his influential book about the prehistory of the Great Plains.1 Most archaeologists are aware that many of their pronouncements about the past reflect their interpretive deficits. In this book I have presented some ideas that attempt to explain the character of the rock art on the central High Plains, but I have assumed that my conclusions represent a starting point for others to build on in future years, and I want to emphasize that my explanations are not “written in stone”; stated another way, they are not nearly as permanent as the images I have written about. Originally I had planned to encompass a larger geographical area and to include information about sites in the panhandles of Texas and Oklahoma. The realization that I had amassed considerable information about the rock art sites at the Piñon Canyon Maneuver Site, where the site density, especially on the Hogback, is quite high, led me to conclude that abundant material was already at hand. I had known there was a high site density in Purgatoire Canyon, but I was not prepared to find such large numbers of sites in the canyon country of southeastern Colorado and northeastern New Mexico. Indeed, there are sufficient sites in those canyons for more than a dozen books.

I

220

8 — THROUGH

A

GLASS, DARKLY

221

It is not as if I had not been forewarned. Mike Waugh and other local rock art enthusiasts had repeatedly told me that sites were plentiful. Indeed, William McGlone and his colleagues wrote several books about the region’s rock art which I had largely overlooked because of their bias toward explaining rock art as a product of Old World Celtic travelers. Because I believe these ideas are farfetched, my search for the meaning underlying the rock art record led me to the use of native ethnographic sources, and these sources have enabled me throughout this book to present some alternative hypotheses for readers to consider as they study the rock art of the region.

ARCHAIC ROCK ART

ON A

REGIONAL SCALE

In Chapter 2, I discussed the oldest rock art on the central High Plains, which is abstract in design. I also included a glimpse of some incised, abstract petroglyphs that may be older than the Archaic. Considerable work needs to be done before these ancient incised figures can be securely assigned to a time period, but they are nonetheless a fascinating part of the High Plains rock art inventory. Of course, we have no idea about the cultural identity of groups in the Archaic Period, but even if we had a vague notion about who made the abstract forms, too much time has elapsed for there to be an ethnographic source explaining the context in which they were created or what they mean. To offset the lack of an “informed” explanation, I have used what Paul Taçon and Christopher Chippindale identify as a “formal” approach to understanding the figures.2 I believe that Early Archaic figures are more self-contained and independent and that it may be possible to determine whether these figures are the oldest rock art in the region by conducting more research at sites at high elevations or in areas of abundant water where there are also archaeological remains of Early Archaic camp sites and game drive features. One of the interesting aspects of Archaic-age rock art is that elsewhere on the High Plains—in the Black Hills of Wyoming and South Dakota, for example— the oldest recognized rock art includes animals in what have been interpreted as communal hunting scenes. Most of the animal figures appear to represent deer, but pronghorns, bighorn sheep, elk, and some bison occur in the panels as well.3 The presence of dog images seems to suggest that this canine species helped to drive animals into nets or along fences made of long, pendant, looping lines. Human hunters carrying spears, clubs, or atlatls appear to be chasing the quadrupeds, although some of these figures are wearing costumes that suggest they are performing a dance or are involved in a ceremony related to the hunt. Very old figures of animals have been found at other localities in Wyoming, as well. At the Legend Rock site near Thermopolis, for instance, the first

222

L AWRENCE L. LOENDORF — THUNDER

AND

HERDS

researchers to work in Wyoming identified a number of realistic animal figures as the oldest in the region—and labeled them as representative of the Early Hunting Tradition—based on their extensive varnish covering. Cation-ratio dates suggest that some of these figures might be Paleo-Indian in age, but even if this assignment is incorrect, most researchers believe that they date to the Early Archaic. My discussion of Archaic Period rock art in Chapter 2 stated that the oldest quadrupeds in the central High Plains appeared in the Late Archaic Period at the Piñon Canyon boulder sites and represent a minor part of the rock art record until the first temporal division (Developmental) of the succeeding Late Prehistoric Period. One of the curious aspects about Developmental Period rock art is the way in which quadrupeds appear so abruptly and in significant numbers. Beginning about 1500 years ago, or about the time that realistic petroglyphs on the northern Plains decrease in number, sites on the central Plains are dominated by quadrupeds. The Big Hands Hunter site, described in detail in Chapter 3, is one such site, although there are many others. The quadrupeds present at these sites, which are sometimes adjacent to an Archaic-age site, suggest that an in-migration of new peoples or a wholesale adoption of a new rock art tradition occurred. There is no evidence in the archaeological record supporting an in-migration, so the introduction of new rock subject matter in the Developmental Period is the more likely explanation. It should be remembered that major changes in material culture occurred at about the same time, primarily the adoption of the bow and arrow, the introduction of cord-roughened ceramics, and the appearance of maize horticulture. The timing of these changes is not well established, and some may date to the Archaic, but there is little question that these innovations became well established in the Developmental Period. Based on the research presented in this book, archaeologists can add the presence of large numbers of quadrupeds in the rock art record to the list of Developmental Period innovations. The question of where the new rock art forms originated is unresolved. It is interesting that similar quadrupeds were made in the Black Hills, but if changes in imagery were associated with dietary innovations, then quadrupeds—like the source of maize—were more likely introduced to the region from the central Plains. The few quadrupeds occurring in the Late Archaic are harbingers of change, preceding an across-the-board transition in the Purgatoire region during the Developmental Period from abstract petroglyphs to quadrupeds in a wide range of sizes and numbers. Some of these have spears or arrows stuck in their backs, and others appear to be connected by undulating lines. Sometimes they occur as isolated figures with no other apparent purpose than to demonstrate what a well-made quadruped looks like. As time passes, anthropomorphs are added to the scenes, first in static postures, but by Apishapa times they are engaged in action-oriented scenes.

8 — THROUGH

A

GLASS, DARKLY

223

PARK PLATEAU VERSUS HIGH PLAINS ROCK ART A relatively rich assemblage of rock art sites is found on the Park Plateau in northern New Mexico, a region utilized from A.D. 400 to 1300 by ancestral Pueblo peoples. Michael Glassow has defined seven chronological phases for this region, two of which—the Ponil phase (A.D. 1100 to 1250) and the Cimarron phase (A.D. 1250 to 1300)—overlap in time with the Developmental Period and the Apishapa phase of southeastern Colorado.4 Much of the archaeological research in this area of the Park Plateau has taken place on the Philmont Scout Ranch, near Cimarron, New Mexico, where rock art studies have been an active part of the program. Sites were first described in a comprehensive report by Dennis Gilpin, a member of the staff at Philmont Scout Ranch in the 1970s, who is now recognized as an extremely capable Southwestern archaeologist. Gilpin identified 22 rock art localities with 161 individual motifs, all pecked into the rock surface except for a horse and an “X” that are incised. There is no evidence that paint was used at any of the sites.5 Gilpin classified the petroglyphs into five different stylistic categories: pecked abstract (undated); pecked pictorial (undated); pecked geometric (A.D. 900 to 1300); Plains Biographic (A.D. 1500 to 1900), and historic Euro-American (A.D. 1541 to present). The pecked abstract style includes lines of dots, wavy lines, undulating lines, squiggles, and mazes, and is likely to be related to the abstract petroglyphs of Archaic age that are found across the region. Finding a fit between the pecked pictorial style and sites on the High Plains is more problematic because of the types of figures that are represented. These include anthropomorphs, a flute player, bear paws, snakes, cranes, thunderbirds, corn plants, and a possible Datura plant. One group of seven anthropomorphs appears to be holding hands in a dance scene. Taken together, these figures have more in common with Pueblo rock art than they do with rock art images typical of Plains cultures. The pecked geometric style, which includes triangles, right-angle lines, and concentric circles that may combined with representational figures like anthropomorphs, is the most common petroglyph type occurring at Philmont Scout Ranch (Figure 8.1). Gilpin points out that this style is “strongly related to Puebloan rock art from the Southwest that began in the Pueblo II period (A.D. 900 to 1100) and lasted until it was supplemented by katcina cult images in the late A.D. 1200s.”6 It is difficult, however, to find similarities between the Philmont rock art inventory and the stylistic categories of the High Plains. Although the pecked geometric style overlaps the Purgatoire Pecked I and II styles chronologically, there are few commonalities between the two. According to Gilpin’s analysis, Philmont’s pecked geometric style contains about 40 percent concentric circles, including those examples incorporated in human or animal figures, but few if any solidly pecked quadrupeds. A typical Purgatoire Pecked I or II site contains about 50 percent quadrupeds and no concentric circle designs.

224

L AWRENCE L. LOENDORF — THUNDER

AND

HERDS

FIGURE 8.1. Concentric circles and other branching motifs at the Philmont Scout Ranch. These figures are contemporaneous with petroglyph panels approximately 100 kilometers to the east that are dominated by quadrupeds. Redrafted figures by Elaine Nimmo from an original illustration by Dennis Gilpin. Used with permission.

Philmont’s pecked pictorial category diverges from rock art on the High Plains by the inclusion of flute players and cranes. Although thunderbirds are common in both areas, the Philmont figures are more stylized and technically different from Purgatoire examples. Bear tracks are common in both areas, but they occur in nearly every rock art tradition, and many of the examples on the High Plains were likely made by Athapaskan groups rather than the Apishapa. The pecked pictorial anthropomorphs are somewhat similar to those on the High Plains, but when they are closely compared, the Philmont examples are more animated and have pointed limbs, unlike the Purgatoire examples. In any case, however, the greatest difference between the two regional rock art inventories is the absence of quadrupeds at Philmont sites. The differences between the rock art of the Park Plateau and that of the High Plains are important for several reasons. Not only is Philmont rock art, as Gilpin noted, closely related to Pueblo II rock art in the Southwest, but perhaps more significantly, the geographical distribution of Philmont styles does not extend onto the High Plains. In fact, the discontinuity in rock art images between the two regions is dramatic. A High Plains rock art site, with more than 250 Purgatoire

8 — THROUGH

A

GLASS, DARKLY

225

Pecked quadrupeds, is located only 100 kilometers northeast of the Philmont Scout Ranch, where a contemporaneous site has more than 70 percent concentric circles. The artifact assemblages and cultural inventories of sites on the Park Plateau and central High Plains have been interpreted by several researchers as indicating that the origin of the Park Plateau tradition can be traced to Plains cultures,7 but I think that the total absence of quadrupeds in Park Plateau rock art reveals that there never was an established relationship between the two regions. The clearly defined boundary between the rock art styles of the Park Plateau and the central High Plains reflects the fact that Apishapa petroglyphs, and those in the preceding Developmental Period, appear to originate among Caddoan groups, while the Park Plateau tradition is Puebloan. Based on the artifacts recovered from Apishapa sites, we should not find this outcome surprising. Small amounts of New Mexico obsidian occur in deposits at Apishapa sites, but there is an almost total absence of painted ceramics, making it clear there was very little contact between the two regions.

A SOCIAL GEOGRAPHY The distinctly different rock art styles of the Park Plateau and the High Plains regions could be a reflection of the principles of “social geography” that Margaret Conkey has proposed to explain the function of rock art in prehistoric societies.8 She suggests that one technique used by groups to identify their territory to outsiders was the placement of symbols on surfaces in the landscape. In some instances, territorial boundaries may simply have been a by-product of the particular rock art symbols that groups left at the places they visited in a seasonal cycle. Rock art researchers have also proposed that figures were intentionally placed at territorial boundaries as a warning to groups entering another group’s territory. An example of the latter use of rock art figures is evident in the Bighorn Basin of Wyoming, where Julie Francis and I found Dinwoody-style petroglyphs with large, often strangely shaped figures in the basin to the west of the Bighorn River, while to the east, the rock art imagery is totally different and includes shield-bearing warriors, bear paws, arrows, and other symbols of power. The placement of power symbols at territorial boundaries was meant to warn others that they were entering a place where the traditional landholders had powerful medicine. We have suggested that the western Dinwoody rock art is oriented to Uto-Aztecan-speaking bands of the Great Basin, and the imagery to the east was created by Siouan or Algonkian-speaking Plains tribes.9 An important implication of the geographical distribution of different rock art figures, at least those near territorial borders, is that they must contain information that outsiders would recognize as symbols belonging to another group. If archaeologists can identify rock art boundaries, then the distribution of motifs

226

L AWRENCE L. LOENDORF — THUNDER

AND

HERDS

becomes an extremely powerful tool. Once a type or style of rock art is reliably associated with an archaeological culture, as I have demonstrated by identifying Purgatoire Pecked I and II styles with the Apishapa phase, the motifs can be used to define the territorial extent of the culture. This last fact has important implications. For example, archaeologists can use the distribution of rock art panels with small human figures associated with large herds of quadrupeds to help them locate Apishapa habitation sites that should occur in the same area. Archaeologists in the American Southwest have followed this strategy for many years as they searched for sites. Earl Morris and Emil Haury, for example, targeted Basketmaker rock art sites in the Four Corners region with the expectation that the Basketmaker archaeological sites they wanted to excavate would be nearby. Although the association between rock art sites and habitation sites is not perfect, this does not negate the fact that a particular rock painting style is an expression of Basketmaker material culture. In many circumstances, if archaeologists can link a rock art style to the archaeological record of the makers of that style, they have discovered a reliable ethnic identifier.

ATHAPASKANS , AGAIN One focus of my ongoing research is to identify a link between the Castle Gardens–style shield-warrior figures and the movements of Athapaskan-speaking peoples. In the process of locating sites with these motifs, I will also be identifying places used by Athapaskan peoples as they migrated from Canada to the American Southwest. In a recent book on Athapaskan migrations, R. G. Matson and Martin Magne use the results of a 25-year archaeological project in the Eagle Lake area of British Columbia to address the problem of Athapaskan movements.10 Their synthetic approach integrates data from the archaeological record with several other resources: a well-controlled ethnoarchaeological study by Linda Burnard-Hogarth among the Chilcotin Athapaskans, traditional ethnographic information, and statistical methods used to analyze lithic assemblages. More importantly, they have developed and used a “parallel direct historical approach” to examine the archaeological records of two adjacent geographic areas that were historically occupied by Salish and Athapaskan groups. By studying two cultural regions, they were able to identify in-migrating Athapaskans by comparing differences, albeit minor, in the lithic technology and stone tool assemblages. Without question, their research presents strong evidence for the archaeological development of Athapaskan groups in British Columbia. The route these Athapaskans took from southern British Columbia to the American Southwest remains an unanswered question, however. Matson and Magne agree that the Castle Gardens–style shield figures in Montana and Wyoming are so similar to those in the Dinetah, the area in New Mexico identified by Navajo

8 — THROUGH

A

GLASS, DARKLY

227

peoples as their place of origin, that an association between the two types of shield-warrior figures appears certain. They further suggest that between A.D. 1400 and 1450—when the Apishapa and Upper Republican cultures retreated from the High Plains to the Plains proper—Athapaskan groups moved through the open territory along the eastern front of the Rocky Mountains on their way to the Plains. Then, at a later time—circa A.D. 1550—groups that became identified as Navajo moved into the Dinetah region of New Mexico. Matson and Magne point to the relatively well-established date of A.D. 1450 for the Tierra Blanca phase, identified in Texas as Apache, as evidence supporting the High Plains entry route. In Chapter 6 on the Protohistoric Period, I presented Ronald Towner’s two models proposing alternative routes that Athapaskan migrants might have followed on their way to the Southwest: the “Early Entry, Mountain Route Refugee Hypothesis Model” and the “Late Entry, High Plains Refugee Hypothesis Model.” Matson and Magne have now introduced a third alternative that could be called the “Early Entry, High Plains Refugee Hypothesis Model.” I have rejected this last model because there is insufficient archaeological evidence to support a proposed early Plains entry route for Athapaskans. Not only are there no Castle Garden–style shield-bearing warriors on the High Plains, but other archaeological remains—identified as Apachean and dated to the 1400s or 1500s—do not exist there. The problem of where Athapaskan groups went after they left British Columbia will require more than the evidence of the rock art record to solve. To make a strong case for an Athapaskan presence in Montana and Wyoming, archaeologists will have to recover—in addition to rock art—the side-notched projectile points, micro-blades, and chipped stone detritus that Matson and Magne have identified as Athapaskan in the Eagle Lake region. Then, if possible, they will have to follow these artifact assemblages to the south. Given our current knowledge of the archaeological record, however, and based on the distribution of Castle Garden–style shield figures, I would argue that Towner’s “Early Entry, Mountain Route Refugee Hypothesis Model” is the more appropriate explanation. While I have yet to discover other Athapaskan artifacts at Castle Gardens–style sites, I think the shield figure itself is a reliable ethnic identifier left behind by proto-Navajo and Apache peoples, and that sites with these figures will continue to be found along the mountain route. The two or three Rio Grande–style sites in the Purgatoire River area are also of considerable interest to researchers trying to account for the movements of Apache groups into the region. It is hard to explain their presence more than 300 kilometers from the core area of Rio Grande–style rock art. It may be significant that most of the imagery at these sites is war-related, suggesting that it was created by a Pueblo war party, but figures such as cornstalks are not as readily associated

228

L AWRENCE L. LOENDORF — THUNDER

AND

HERDS

with conflict. One possibility is that the sites were created in the context of an attempt to colonize the region after the Apishapa abandonment of the area. It may also be time to rethink some of the rock paintings currently classified as exemplars of the Purgatoire Painted style, which may actually have been created by Pueblo and Apache peoples. I believe that we, as researchers, place too many paintings within the Purgatoire Painted taxon and that a more refined study is needed to sort out those that belong in other categories. For example, a number of painted mask-like images that resemble Pueblo designs from central New Mexico are grouped with Purgatoire Painted figures. At present these red and black figures have only been found at McNees Crossing, north of Clayton, New Mexico, about 90 kilometers south of the primary Purgatoire region, suggesting they might also be found in northern localities. McNees Crossing contains a group of anthropomorphs with long, linear bodies, short crossed arms, and round heads. These figures also appear stylistically similar to paintings along the Rio Grande to the south. To complicate the picture further, several painted quadrupeds at McNees Crossing are very similar to those at the Red-tail rockshelter and the Game Drive site, perhaps indicating that the McNees Crossing site represents the southernmost extent of the Purgatoire Painted style. The presence of Rio Grande–style motifs in the north might lead to the suggestion that at some point, probably after A.D. 1400 when the Apishapa abandoned the region, Pueblo groups ventured onto the High Plains. There is, however, very little archaeological evidence, such as painted ceramic sherds, that might support a Pueblo presence in the Purgatoire area. At a small number of sites, some anthropomorphs with very small heads and narrow waists that resemble Apache anthropomorphs from southern New Mexico have been included in the Purgatoire Painted style (Figure 8.2). The red pigments often used to paint these figures perhaps account for their grouping within the Purgatoire Painted category. The clearly Apachean anthropomorphs in the southern New Mexico sites, for which not only red but yellow and white pigments were used, are frequently depicted with horned headgear. Some are shown on horseback, and several scenes feature dancers that may represent or be related to gans figures. Variability within the Purgatoire Painted style itself makes it difficult to say with certainty whether particular features are part of the style or actually belong to either Pueblo or Apache painted rock art types. The three sites described in the Purgatoire Painted chapter exemplify the complexity within the style. Images at the Rock Crossing site consist primarily of abstract designs, while at the Game Drive site quadrupeds and small humans predominate. The Bear Dance site is probably a Caddoan ceremonial site, yet all three rock art assemblages represent variations of the Purgatoire Painted style. The heterogeneity of the style is therefore responsible for the difficulty researchers experience in identifying painted figures that do not belong in the style.

8 — THROUGH

A

GLASS, DARKLY

229

FIGURE 8.2. Suspected Apache painted figures that are often confused with Purgatoire Painted images. Drawing by Elaine Nimmo from a photograph by Lawrence Loendorf.

APISHAPA ROCK ART Another unanswered question—which is not unique to the rock art of the central High Plains—is why did Apishapa groups make rock art in the first place? Information relevant to this subject is found in a recently published paper by James Keyser and David Whitley in which they discuss two contrasting meanings underlying very similar rock art figures. The authors propose that some Columbia Plateau animal figures with spears or arrows in their backs represent hunting magic, whereas in California and Nevada, animals (often bighorn sheep) with spears in their backs are better interpreted as metaphors for the death that a shaman experiences when entering a trance state.11 They reason that the scenes associated with shamans would include nonrealistic imagery unique to the individual experiencing the trance. On the Columbia Plateau, however, rock art scenes are realistic and differ very little in content. These recurrent images would therefore appear to be symbols of sympathetic magic intended to produce a successful hunting outcome. Keyser and Whitley base their contentions on ethnographic sources describing tribes in California and Nevada as lacking knowledge of the use of rock art as hunting magic, whereas ethnographies of Columbia Plateau tribes contain many accounts of events in which game animals were depicted and then magically killed in order to ensure a positive outcome of a hunt. Depictions of High Plains animals with spears or arrows in them or those being driven toward a net are more problematic. While there is relatively strong ethnographic support for native hunters in the Great Lakes region using rock art sites for sympathetic magic (discussed in Chapter 4), there are no similar ethnographic resources for the central High Plains tribes. Based on my research, however, I believe the animal drive scenes in High Plains rock art are unquestionably part of a hunting complex. Whether they are the product of a shaman’s vision or were made as part of sympathetic magic rituals, however, remains unresolved.

230

L AWRENCE L. LOENDORF — THUNDER

AND

HERDS

Fortunately, archaeological and ethnographic information regarding ceremonies associated with deer and buffalo hunting on the High Plains is substantial. Unusual evidence was uncovered in the 1970s at the Crenshaw site in southwestern Arkansas, where, adjacent to a structure, archaeologists excavated a pile of 2042 deer antlers from mature, robust bucks. Only a single spike from a yearling buck was among the collection of antlers from older deer. Furthermore, the excavator, Frank Schambach, believed that at least one-third of the original assemblage had been destroyed by recent agricultural plowing.12 This estimate means that, based on their pristine condition, a total of 3000 deer antlers had been kept under a roof next to what appeared to be a ceremonial structure. The Crenshaw site is proto-Caddoan and is associated with social groups that very likely spoke a language shared with the makers of the central High Plains petroglyphs. Another relevant fact is that the Crenshaw site is located on the Red River, a significant part of whose headwaters rises in the Texas panhandle. Ancient trails, later used in the Historic Period to drive cattle, connect Texas and Colorado, and their presence supports the suggestion that the makers of rock art on the central High Plains had knowledge of and participated in a regional deer ceremonialism that was expressed in a variety of ways in different places. The time-depth of Caddoan deer-hunting rituals is further attested to by Spanish accounts of the Hasinai, a Caddoan people from eastern Texas and Arkansas: Before deer hunts, they observed a ceremonial to assure success of the expedition. The hollowed-out head of a deer, complete with neck and antlers, was placed on a post in the house and prayers were made to the caddi ayo to allow the deer to fall into the hands of the hunter. . . . When . . . the ceremonial was completed, the deer head was removed and placed at the door of the house and, taking identical heads with which to camouflage themselves, the hunters, their naked bodies covered with white dirt, departed for the chase.13 A review of the evidence of deer ceremonialism among Caddoan and protoCaddoan peoples begins, then, with deer images at several central High Plains sites and the roughly contemporary assemblage of deer antlers at the Crenshaw site, and is ongoing in the sixteenth and seventeenth centuries, when Spanish records describe Hasinai hunters praying before deer heads prior to a hunt. Even more recent ethnographic accounts of pre-hunt rituals to ensure success provide details of a Skidi Pawnee ritual associated with a circle or semicircle of buffalo skulls.14 The account by Good Heart of the use of a stone buffalo as a prayer station, which I presented in the discussion of Purgatoire Pecked II sites, is another example of this tradition. The preceding ethnographic evidence is supported by the discovery of archaeological sites associated with some of the game drive rock art panels. The rock drive lines and the natural funnel-shaped settings are also strong evidence

8 — THROUGH

A

GLASS, DARKLY

231

that rock art was an important part of a successful drive. As I reported briefly in previous chapters, my team and I have begun to excavate in the rockshelters associated with drive sites and have found antelope bones that indicate the animals were taken from the kill area to a nearby location for butchering. The presence in such a context of the hunt-related rock art imagery that I have described previously—incised nets both underneath and above quadruped figures, animal figures caught in such nets, and the damage to figures caused by overstriking which I believe represents places where arrows were shot at petroglyph and pictograph animals or where the figures were jabbed with spears as a way to magically kill them—suggests that at least some of the rock art created by Apishapa people relates to the techniques they devised to capture the fleet-footed, thundering herds of the central High Plains on which their subsistence depended.

A GLIMPSE

OF THE

FUTURE

The verse from I Corinthians 13 with which I began this chapter concludes on a hopeful note: the vision seen darkly, perhaps through the whiskey bottle, ultimately will be replaced by direct knowledge. Archaeologists may bemoan the fact that they will never know the past directly, for theirs is an inferential discipline, but the good news is that more and more of the past is becoming clear through arguments that link different domains of knowledge. It could be said that the archaeological record becomes more meaningful as the number of “glasses” used to view it increases. Some of these lenses relate to the spatial and temporal distribution of the materials we study. Others refract scenes and activities from ethnographic sources that appear relevant to, for instance, the interpretation of rock art images. It is only by combining insights from these various intellectual domains that we will develop a comprehensive picture of past life on the High Plains, and the place in it of rock art.

NOTES 1 Wedel 1961 2 Taçon and Chippindale 1998. 3 Sundstrom 2004:55; Tratebas 2000. 4 Glassow 1980. 5 Gilpin 2001. 6 Gilpin 2001:68. 7 Baugh 1994; Dorshow 1997. 8 Conkey 1984. 9 Francis and Loendorf 2002. 10 Matson and Magne 2007. 11 Keyser and Whitley 2006. 12 Schambach 1977. 13 Griffith 1954:115–16, quoted from Scott and Jackson 1998:6. 14 Grinnell 1889:270.

REFERENCES

Abel, Brent. 1994. Proposed Investigations at Two Western Archaic Petroglyph Sites near Pecos, New Mexico. In American Indian Rock Art, vol. 19, ed. Frank G. Bock, 13–18. San Miguel, CA: American Rock Art Research Association. Academy of Achievement. 2006. Interview with Donald C. Johanson, Ph.D. Electronic document available at www.achievement.org/autodoc/page/joh1int5. Aikens, Melvin. 1970. Hogup Cave. Anthropological Papers, no. 93. Salt Lake City: University of Utah Press. Albers, Patricia. 2003. The Home of the Bison: An Ethnographic and Ethnohistorical Study of Traditional Cultural Affiliations to Wind Cave National Park. Report on file with the National Park Service and the Department of American Indian Studies, University of Minnesota. Andrefsky, William, Marilyn Bender, John Benko, and Judy Michaelsen. 1990. Test Excavations in the Piñon Canyon Maneuver Site, Southeastern Colorado. Two-volume report on file with the United States Army, Fort Carson, CO. Anell, Bengt. 1969. Running Down and Driving of Game in North America. Lund, Sweden: Studia Ethnographica Upsaliena 30. Anonymous. 1904. Los Angeles Times, October 5. Anonymous. 2007. Lost and Found. Journal of California and Great Basin Anthropology 27(1):79–81. Baldwin, Stuart. 1997. Apacheans Bearing Gifts: Prehispanic Influence on the Pueblo Indians. The Arizona Archaeologist, 29. Phoenix: Arizona Archaeological Society. Barton, C. Michael, Geoffrey Clark, David Yessner, and Georges Pearson. 2004. The Settlement of the American Continents: A Multidisciplinary Approach to Human Biogeography. Tucson: University of Arizona Press. Baugh, Timothy. 1994. Holocene Adaptations in the Southern High Plains. In Plains Indians, A.D. 500–1500: The Archaeological Past of Historic Groups, ed. K. H. Schlesier, 264–89. Norman: University of Oklahoma Press. Begole, Robert. 1985. A Game of Skill? Pacific Coast Archaeological Society Quarterly 21(4):60–62. Bement, Leland. 1999. Bison Hunting at Cooper Site: Where Lightning Bolts Drew Thundering Herds. Norman: University of Oklahoma Press. Benedict, James. 1979. Getting Away from It All: A Study of Man and the Two-Drought Altithermal. Southwestern Lore 45(3):1–12.

232

REFERENCES

233

Benedict, James, and B. L. Olson. 1978. The Mount Albion Complex: A Study of Prehistoric Man and the Altithermal. Research Report No 1. Ward, CO: Center for Mountain Archaeology. Berlo, Janet. 2000. Spirit Beings and Sun Dancers: Black Hawk’s Vision of the Lakota World. New York: George Braziller, Inc. Binford, Lewis R. 1983. Working at Archaeology. New York: Academic Press. Black, Kevin. 1991. Archaic Continuity in the Colorado Rockies: The Mountain Tradition. Plains Anthropologist 36(133):1–29. Blakeslee, Donald J. 2003. A Meteor, A Mirror, A Metaphor. The Wisconsin Archaeologist 84(1 & 2):99–106. Blakeslee, Donald, Robert Blasing, and Hector Garcia. 1986. Along the Pawnee Trail: Cultural Resource Survey and Testing at Wilson Lake, Kansas. Report on file with the United States Army Engineer District, Kansas City, MO. Blasing, Robert. 1993. A Possible Location for the “Garden of the Gods” Described in Pawnee Stories. Paper presented at the Flint Hills Conference, Wichita, KS. Blythe, Jeffrey. 2007.Our Footprints Are There: Report of Native American Consultation to Identify Cultural Properties and Sacred Sites on Lands. Report on file with the United States Army, Fort Carson, CO. Bock, Frank G., and A. J. Bock. 1992. Final Report of the Recording of Three Petroglyph Sites on the Glorieta Mesa, San Miguel County, New Mexico. Manuscript on file, Santa Fe National Forest, Santa Fe, NM. ——. 1993. Addendum to Final Report of the Recording of Three Petroglyph Sites on the Glorieta Mesa, San Miguel County, New Mexico. Manuscript on file, Santa Fe National Forest, Santa Fe, NM. ——. 1994. Three Sites on Glorieta Mesa, New Mexico, Show Evidence of Western Archaic Tradition Rock Art. In Rock Art Papers, vol. 11, ed. Ken Hedges, 117–30. San Diego Museum of Man Papers 31. San Diego: San Diego Museum of Man. Bourke, John. 1892. The Medicine-Men of the Apache. Ninth Annual Report of the Bureau of American Ethnology, 443–617. Washington, DC. Boyd, Carolyn. 2003. Rock Art of the Lower Pecos. College Station: Texas A & M University Press. Brugge, David. 1983. Apacheans in Plains Culture History. Bulletin of the Oklahoma Anthropological Society 32:107–14. Buckles, William G. 1964. An Analysis of Primitive Rock Art at Medicine Creek Cave, Wyoming, and Its Cultural and Chronological Relationships to the Prehistory of the Plains. Master’s thesis, Department of Anthropology, University of Colorado, Boulder. ——. 1971. Part Five: Rock Art. In The Uncompahgre Complex: Historic Ute Archaeology and Prehistoric Archaeology on the Uncompahgre Plateau in West Central Colorado. Ph.D. thesis, Department of Anthropology, University of Colorado, Boulder. ——. 1974. Petroglyphs at Site 5BN72. Southwestern Lore 40(3 & 4):72–74. ——. 1980. Radiocarbon Dates of Deposits of Piney Creek Alluvium in Southeastern Colorado. Southwestern Lore (3):18–22. ——. 1982. Dated Plains Rock Art Sites. Paper presented at the Annual Meeting of the Colorado Archaeological Society, October 9, 1982. Manuscript in the archives of the Colorado Rock Art Association.

234

L AWRENCE L. LOENDORF — THUNDER

AND

HERDS

——. 1989. Petroglyphic Research and Ogam in Southeastern Colorado: Strategies for Resolving Controversies. In Rock Art of the Western Canyons, ed. Jane Day, Paul Friedman, and Marcia Tate, 113–55. Publication of the Denver Museum of Natural History and the Colorado Archaeological Society, Denver. Burton, Robert J. 1971. The Pictographs and Petroglyphs of Dinosaur National Monument. Master’s thesis, Department of Anthropology, University of Colorado, Boulder. Callahan, Kevin. 2000. Pica, Geophagy, and Rock Art: Ingestion of Rock Powder and Clay by Humans and Its Implications for the Production of Some Rock Art on a Global Basis. Paper read at the 65th Society for American Archaeology meetings, Philadelphia, PA. Callaway, D. G., J. C. Janetski, and O. C. Stewart. 1986. Ute. In Handbook of North American Indians, Vol. 11: Great Basin, ed. Warren L. D’Azevedo, 336–67. Washington, DC: Smithsonian Institution. Campbell, Robert. 1969. Dating Prehistoric Rock Art of Southeastern Colorado. Southwestern Lore 35(1):1–10. ——. 1976. The Panhandle Aspect of the Chaquaqua Plateau. Graduate Studies, no. 11. Lubbock: Texas Tech Press. Carrillo, Richard. 1999. A General Summary of the Ethnohistory and History of the Purgatoire and Arkansas Valley Regions in Southeastern Colorado, 1640s–1870s. Manuscript in the possession of the author, La Junta, CO. Cassells, E. Steve. 1992. A History of Colorado Archaeology 1. In The State of Colorado Archaeology, ed. by Phillip Duke and Gary Mattock, 4–34. Colorado Archaeological Society Memoir 5, Denver. ——. 1997. The Archaeology of Colorado. Revised edition. Boulder, CO: Johnson Books. Chamberlain, Von Del. 1982.When the Stars Came Down to Earth: Cosmology of the Skiri Pawnee Indians of North America. Anthropological Papers, no. 56. Los Altos, CA: Ballena Press. Chomko, Stephen, Steve Devore, and Lawrence L. Loendorf. 1990. Apishapa Phase Research at the Piñon Canyon Maneuver Site, Southeastern Colorado. Paper presented at the 48th Plains Conference, Oklahoma City, OK. Church, Minette, and Pamela Cowen. 2003. Evaluative Testing of the Bent Canyon Stage Station on the Piñon Canyon Maneuver Site, at Las Animas County, Colorado. Report on file with the United States Army, Fort Carson, CO. Clark, LaVerne.1983.They Sang for Horses: The Impact of the Horse on Apache and Navajo Folklore. Tucson: University of Arizona Press. Cole, Sally. 1985. Interim Rock Art Report for the Fort Carson-Piñon Canyon (Las Animas County, Colorado) Archaeological Project: 1983 Field Season. In The Fort Carson-Piñon Canyon Project: Interim Management Report, compiled by Thomas Pozorski and Mark Guthrie, 249–305. Denver: Archaeological Research Institute, University of Denver. ——. 1987. An Analysis of the Prehistoric and Historic Rock Art of West-Central Colorado. Cultural Resource Series 21, Bureau of Land Management, Grand Junction, CO. ——. 1990. Legacy on Stone: Rock Art of the Colorado Plateau and Four Corners Region. Boulder, CO: Johnson Books.

REFERENCES

235

Collins, M. B., T. R. Hester, and P. J. Headrick. 1992. Engraved Cobbles from the Gault Site, Central Texas. Current Research in the Pleistocene 9:3–4. Colorado Springs Free Press, Sunday, June 20, 1965, “LBJ Makes Declaration as Arkansas River Rages,” p. 1. The Compact Edition of the Oxford English Dictionary. 1971. Oxford: Oxford University Press. Conkey, Margaret. 1984. To Find Ourselves: Art and Social Geography of Prehistoric Hunter Gatherers. In Past and Present in Hunter Gatherer Studies, ed. Carmel Schrire, 253–76. Orlando, FL: Academic Press. Conner, Stuart. 1982. Archaeology of the Crow Indian Vision Quest. Archaeology in Montana 23(3):85–127. Conway, Thor, and Julie Conway. 1990. Spirits on Stone: The Agawa Pictographs. San Luis Obispo, CA: Heritage Discoveries. Cummings, Linda Scott. 1990. Pollen Report for 5LA5586, Piñon Canyon Maneuver Site. Manuscript in the possession of Lawrence L. Loendorf. Dean, J. Claire. 1991. Rock Art Conservation and the Pinon Canyon Maneuver Site, Colorado, Progress Report Number 1. Report on file with the Cultural Resource Program, Fort Carson, CO. Dorn, Ronald I. 1989. Cation-Ratio Dating. In Nine Rock Art Sites in the Pinon Canyon Maneuver Site, Southeastern Colorado, ed. Lawrence Loendorf, 128–37. Contribution 248. Grand Forks: Department of Anthropology, University of North Dakota. ——. 1993. Letter from Ronald I. Dorn to Frank and A. J. Bock dated December 21, 1993. Attached to report number 92–10–125–D. On file with Santa Fe National Forest, Pecos, New Mexico. ——. 1994. Letter from Ronald I. Dorn to Brent Abel dated August 16, 1994. On file with Santa Fe National Forest, Pecos, New Mexico. ——. 1996. A Change of Perception. La Pintura 23(2):10–11. ——. 2001. Chronometric Techniques: Engravings. In Handbook of Rock Art Research, ed. David Whitley, 167–89. Walnut Creek, CA: AltaMira Press. Dorn, Ronald I., William McGlone, and Phillip Leonard. 1990. Age Determination of the Petroglyphs of Southeast Colorado. Southwestern Lore 56(2):21–36. Dorsey, George A. 1904a. Traditions of the Skidi Pawnee. Boston and New York: Houghton, Mifflin and Company. ——. 1904b. The Mythology of the Wichita. Washington, DC: Carnegie Institution. Dorshow, Wetherbee. 1997. Ancho Canyon Chronology. In Cultural Definition in the Southern Park Plateau of Northeast New Mexico: The Ancho Canyon Project, ed. J. V. Biella and W. B. Dorshow, 911–47. Prepared for Pittsburg and Midway Coal Mining Company by Southwest Archaeological Consultants, Inc., Santa Fe, NM. Echo-Hawk, Roger. 2000. Ancient History in the New World: Integrating Oral Traditions and the Archaeological Record in Deep Time. American Antiquity 65(2): 267–90. Eddy, John A. 1974. Astronomical Alignments of the Bighorn Medicine Wheel. Science 184(4141):1035–43. Edington, Thomas S., and John Taylor. 1998. The Battle of Glorieta Pass: A Gettysburg in the West, March 26–28, 1862. Albuquerque: University of New Mexico Press.

236

L AWRENCE L. LOENDORF — THUNDER

AND

HERDS

Ewers, John C. 1939. Plains Indian Painting: A Description of an Aboriginal American Art. Palo Alto, CA: Stanford University Press. ——. 1955. The Horse in Blackfoot Indian Culture: With Comparative Material from Western Tribes. Bureau of American Ethnology Bulletin 159. Washington, DC: Smithsonian Institution, Government Printing Office. Faris, Peter. 1995. Petroglyph Chronology in Southeast Colorado. Southwestern Lore (1):7–35. Feyhl, Kenneth. 1980. Tool Grooves: A Challenge. Archaeology in Montana 21:1–31. Francis, Julie E. 2001. Style and Classification. In Handbook of Rock Art Research, ed. David S. Whitley, 221–44. Walnut Creek, CA: AltaMira Press. Francis, Julie E., and Lawrence L. Loendorf. 2002. Ancient Visions: Petroglyphs and Pictographs of the Wind River and Bighorn Country, Wyoming and Montana. Salt Lake City: University of Utah Press. Friedman, Paul. 1985. Final Report of History and Oral History. Studies of the Fort Carson Piñon Canyon Maneuver Site Area, Las Animas County, Colorado. Copies available through the Cultural Resource Program at the United States Army Fort Carson Military Base. Frison, George. 1991. Prehistoric Hunters of the High Plains. San Diego: Academic Press. Gill, George. 1983. Navajo Views of Their Origin. In Handbook of North American Indians, Vol. 10: Southwest, ed. A. Ortiz, 502–5. Washington, DC: Smithsonian Institution. Gilmore, K., M. Tate, M. Chenault, B. Clark, T. McBride, and M. Wood. 1999. Colorado Prehistory: A Context for the Platte River Basin. Denver: Colorado Council for Professional Archaeologists. Gilpin, Dennis. 2001. The Rock Art of Philmont. Report in the possession of Dennis Gilpin, Flagstaff, AZ. Glassow, Michael. 1980. Prehistoric Agricultural Development in the Northern Southwest: A Study in Changing Patterns of Land Use. Anthropological Papers, no. 16. Socorro, NM: Ballena Press. Goodwin, Grenville. 1929–1939. Papers held at the Arizona State Museum Archives, University of Arizona, Tucson. Greer, Mavis. 1995. Archaeological Analysis of Rock Art Sites in the Smith River Drainage of Central Montana. Ph.D. thesis, Department of Anthropology, University of Missouri, Columbia. Griffith, W. J. 1954. The Hasinai Indians of East Texas as Seen by Europeans, 1687–1772. Philological and Documentary Studies 2(3). Middle American Research Institute. New Orleans, LA: Tulane University. Grinnell, George B. 1889. Pawnee Hero Stories and Folk Tales. New York: Forest and Stream Publishing Company. Gunnerson, J., and D. Gunnerson. 1971. Apachean Culture: A Study in Unity and Diversity. In Apachean Culture and History and Ethnology, K. Basso and M. Opler, 7–17. Anthropological Papers of the University of Arizona, no. 21. Tucson: University of Arizona Press. Halasi, Judith, David Stuart, and Bruce Rippeteau. 1981. Preliminary Report of Investigations at Hackberry Springs (5LA1115 and 5BA31). Southwestern Lore 47(1): 1–11.

REFERENCES

237

Hamalainen, Pekka. 1998. The Western Comanche Trade Center: Rethinking the Plains Indian Trade System. The Western Historical Quarterly 29(4):485–513. Harrod, Howard. 2000. The Animals Came Dancing: Native American Sacred Ecology and Animal Kinship. Tucson: University of Arizona Press. Hartley, Ralph, and Anne Wolley Vawser. 2003. Rockshelters, Rockart, and Grinding Activity: A Preliminary Assessment of Relationships in Picket Wire Canyonlands, Comanche National Grasslands. Midwest Archeological Center, National Park Service, Lincoln, NE. Hedges, Ken. 1982. Great Basin Rock Art Styles: A Revisionist View. American Indian Rock Art, vols. 7–8, ed. Frank G. Bock, 205–11. El Toro, CA: American Rock Art Research Association. Heizer, Robert, and Martin Baumhoff. 1962. Prehistoric Rock Art of Nevada and Eastern California. Berkeley and Los Angeles: University of California Press. Heywood, John. 1906. A Dialogue of Effectual Proverbs in the English Tongue Concerning Marriage. Edited by John S. Farmer. London: Gibbings and Co. Facsimile copy available from Google Booksearch. Hoijer, Harry, and Morris E. Opler. 1938. Chiricahua and Mescalero Apache Texts. The University of Chicago Publications in Anthropology, Linguistic Series. Chicago: University of Chicago Press. Hotz, Gottfried. 1970. Indian Skin Paintings from the American Southwest. Norman: University of Oklahoma Press. Hoy, Judith. 1969. Petroglyph Boulders in Phillips County, Montana. Archaeology in Montana 10(3):1–19. Irwin, Lee. 1996. The Dream Seekers: Native American Visionary Traditions of the Great Plains. Norman: University of Oklahoma Press. James, S. R. 1983. An Early Incised Stone from Danger Cave, Utah. Journal of California and Great Basin Anthropology 5:247–52. Jansen Thomas, Peter Forster, Marsha A. Levine, Hardy Oilke, Matthew Hurles, Colin Renfrew, Jürgen Webe, and Klaus Olek. 2002. Mitochondrial DNA and the Origins of the Domestic Horse. Proceedings of the National Academy of Sciences 99(16):10905–10. Johnson, Boma. 1986. Earth Figures of the Lower Colorado and Gila River Deserts: A Functional Analysis. The Arizona Archaeologist 20. Phoenix: Arizona Archaeological Society. Kalasz, Stephen, Mark Mitchell, and Christian Zier. 1999. Late Prehistoric Stage. In Colorado Prehistory: A Context for the Arkansas River Basin, ed. Christian Zier and Stephen Kalasz, 141–263. Denver: Colorado Council of Professional Archaeologists. Kavanagh, T. W. 1996. The Comanches: A History 1706–1875. Lincoln and London: University of Nebraska Press. ——. 2001. Comanche. In Handbook of North American Indians, Vol. 13(2): Plains, ed. R. J. DeMallie, 886–906. Washington, DC: Smithsonian Institution. Keyser, James D. 1987. A Lexicon for Historic Plains Indian Rock Art: Increasing Interpretive Potential. Plains Anthropologist 32(115):43–71. ——. 2004. Art of the Warriors: Rock Art of the American Plains. Salt Lake City: University of Utah Press. ——. 2004. Bear Coming Out. Wyoming Archaeologist 48(2):35–42.

238

L AWRENCE L. LOENDORF — THUNDER

AND

HERDS

Keyser, James D., and Michael Klassen. 2001. Plains Indian Rock Art. Seattle: University of Washington Press. Keyser, James D., and Mark Mitchell. 2000. Red Rock Ledge: Plains Biographic Rock Art in the Picketwire Canyonlands, Southeastern Colorado. Southwestern Lore 66(20): 22–37. Keyser, James D., and David S. Whitley. 2006. Sympathetic Magic in Western North American Rock Art. American Antiquity 71(1):3–26. King, Thomas. 1978. A Petroglyph Assemblage from Cerrito de Casacabeles. In Seven Rock Art Sites in Baja California, ed. C. W. Meighan and V. L. Pontoni, 124–77. Socorro, NM: Ballena Press. Klassen, Michael. 2005. “Aisinai’pi” (Writing-on-Stone) in Traditional, Anthropological, and Popular Thought. In Discovering North American Rock Art, ed. Lawrence L. Loendorf, David S. Whitley, and Christopher Chippindale, 15–50. Tucson: University of Arizona Press. The Lamar-Tri-State Daily News, 25th Anniversary Supplement, Tuesday, June 17, 1990, “Holly Remembers the ’65 Flood,” p. 7. Las Animas Leader. August 14. 1874, “Ursa Major: Or the Great Bear of the Purgatoire,” p. 1. Lewis-Williams, J. David, and Thomas Dowson. 1988. The Signs of All Times: Entoptic Phenomena and Upper Paleolithic Art. Current Anthropology 29:201–49. Lintz, Christopher. 1999. Hal Chase, the Snake Blakeslee Site, and the Archaeology of Southeastern Colorado. Southwestern Lore (2):5–31. Lockley, Martin, Barbara Fillmore, and Lori Marquardt. 1997. Dinosaur Lake: The Story of the Purgatoire Valley Dinosaur Tracksite Area. Special Publication 40. Denver: Colorado Geological Society. Loendorf, Lawrence L. 1989. Nine Rock Art Sites in the Piñon Canyon Maneuver Site, Southeastern Colorado. Contribution 248. Department of Anthropology, University of North Dakota, Grand Forks. ——. 1990. A Dated Rock Art Panel of Shield Bearing Warriors in South Central Montana. Plains Anthropologist 35(127):45–54. ——. 1991. Cation-ratio Varnish Dating and Petroglyph Chronology in Southeastern Colorado. Antiquity 65(247):246–55. ——. 1992. The Zookeeper Petroglyph Site, 5LA5993. Southwestern Lore 58(1):17–28. ——. 2001. Rock Art Recording. In Handbook of Rock Art Research, ed. David S. Whitley, 55–79. Walnut Creek, CA: AltaMira Press. ——. 2004. Rock Art and Southward Moving Athapaskans. In Ancient and Historic Lifeways in North America’s Rocky Mountains. Proceedings of the 2003 Rocky Mountain Conference, Estes Park, CO, ed. Robert Brunswig and William Butler, 94–109. Copies available through the Department of Anthropology, University of Northern Colorado, Greeley. Loendorf, Lawrence L., and Stuart Conner. 2004. The Painted Buffalo in Teton Jackson Cave, Montana. Paper presented at the Plains Anthropological Conference, Billings, MT, October 13–15, 2004. Loendorf, Lawrence L., and David D. Kuehn. 1991. 1989 Rock Art Research—Piñon Canyon Maneuver Site, Southeastern Colorado. Contribution 258. Grand Forks: Department of Anthropology, University of North Dakota.

REFERENCES

239

Loendorf, Lawrence L., and Linda Olson. 2003. The Tolar Petroglyph Site. In American Indian Rock Art, vol. 29, ed. Alanah Woody and Joseph T. O’Conner, 1–10. Tucson, AZ: American Rock Art Research Association. Loendorf, Lawrence L., Jeani Borchert, and Duane Klinner. 1996. Archeological Investigations at Ceramic Stage Site in the Piñon Canyon Maneuver Site, Colorado. Contribution 308. Grand Forks: Department of Anthropology, University of North Dakota. Loendorf, Lawrence L., Cynthia Kordecki, and Michael Gregg. 1988. Preliminary Report, Piñon Canyon Maneuver Site Rock Art Project. Report on file with the United States Army, Fort Carson, CO. MacGregor, J. G. 1966. Peter Fidler: Canada’s Forgotten Surveyor, 1769–1822. Montreal: McClelland and Stewart Limited. Mallery, Garrick. 1972. Picture Writing of the American Indians, vol. 2. New York: Dover Publications, Inc. Matson, R. G. and Martin Magne. 2007. Athapaskan Migrations: The Archaeology of Eagle Lake, British Columbia. Tucson: University of Arizona Press. McFaul Michael. 1991. Geoarcheological Interpretation of Alluvial Sedimentation. In 1989 Rock Art Research—Piñon Canyon Maneuver Site, Southeastern Colorado, ed. Lawrence L. Loendorf and D. D. Kuehn, 25–37. Contribution 258. Grand Forks: Department of Anthropology, University of North Dakota. McGlone, Bill, Ted Barker, and Phil Leonard. 1994. Petroglyphs of Southeast Colorado and the Oklahoma Panhandle. Kamas, UT: Mithras, Inc. McGlone, Bill, Phil Leonard, and Ted Barker. 1999. Archaeoastronomy of Southeast Colorado and the Oklahoma Panhandle. Kamas, UT: Mithras, Inc. Meltzer, David. 1999. Human Responses to Middle Holocene (Altithermal) Climates on the North American Great Plains. Quaternary Research 52:404–16. Miller, Alfred Jacob. 1973. Braves and Buffalo: Plains Indian Life in 1837. Toronto: University of Toronto Press. Mitchell, Mark. 2002. Protohistoric and Early Historic Period Rock Art Resources of Picture Canyon, Comanche National Grassland, Baca County, Colorado. United States Department of Agriculture, Forest Service Cultural Resources Series 1817. Report on file with Comanche National Grassland, La Junta, CO. ——. 2004a. Rio Grande Tradition Rock Art in the Upper Arkansas River Basin, Southeastern Colorado. In Ancient and Historic Lifeways in North America’s Rocky Mountains. Proceedings of the 2003 Rocky Mountain Conference, Estes Park, Colorado, ed. Robert Brunswig and William Butler, 110–18. Copies available through the Department of Anthropology, University of Northern Colorado, Greeley, CO. ——. 2004b. Tracing Comanche History: Eighteenth-Century Rock Art Depictions of Leather-Armoured Horses from the Arkansas River Basin, Southeastern Colorado, USA. Antiquity 78(299):115–26. Mobley, Charles. 1981. Archaic Hunter-Gatherer Settlement in Northeastern New Mexico. Ph.D. thesis, Department of Anthropology, Southern Methodist University, Dallas, TX. Murie, James. 1984. Ceremonies of the Pawnee, edited by Douglas R. Parks. Originally published as Smithsonian Contribution to Anthropology, no. 27 (1981). Lincoln: University of Nebraska Press.

240

L AWRENCE L. LOENDORF — THUNDER

AND

HERDS

Nabokov, Peter. 2006. Where the Lightning Strikes. London: Viking Penguin. Newman, Bonita, and Lawrence L. Loendorf. 2005. Portable X-Ray Fluorescence Analysis of Rock Art Pigments. Plains Anthropologist 50(195):277–83. O’Brien, Patricia. 1986. Prehistoric Evidence for Pawnee Cosmology. American Anthropologist 88(4):939–46. ——. 1991. Evidence for the Antiquity of Women’s Roles in Pawnee Society. In Approaches to Gender Processes on the Great Plains, ed. Marcel Kornfeld, 51–68. Plains Anthropologist Memoir 26, vol. 36, no. 134. O’Brien, Patricia, and Diane Post. 1988. Speculation about Bobwhite Quail and Pawnee Religion. Plains Anthropologist (33):489–504. Ooten, Susan. 1992. Ancient Dreams and Stardust Memories: The Amateurs’ Impact on Colorado Archaeology. In The State of Colorado Archaeology, ed. Phillip Duke and Gary Matlock, 109–48. Colorado Archaeological Society Memoir no. 5, Denver. Opler, Morris. 1935. The Concept of Supernatural Power among the Chiricahua and Mescalero Apache. American Anthropologist 37(1):65–70. ——. 1946. The Creative Role of Shamanism in Mescalero Apache Mythology. Journal of American Folklore 59(233):268–81. ——. 1965. An Apache Lifeway. New York: Cooper Square Publishers. ——. 1983. The Apache Culture Pattern and Its Origins. In Handbook of North American Indians, Vol. 10: Southwest, ed. A. Ortiz, 368–92. Washington, DC: Smithsonian Institution. Owens, Mark. 2006. Edge-Ground Cobble Tools of the Pinon Canyon Maneuver Site. Southwestern Lore 72(3):1–27. Penn, B. S., and Lindsey, D. A. 1996. Tertiary Igneous Rocks and Laramide Structure and Stratigraphy of the Spanish Peaks Region, South-Central Colorado: Road Log and Description from Walsenburg to La Veta (First Day) and La Veta to Aguilar (Second Day), Colorado. U.S. Geological Survey Open-File Report 96–4. Phillips, John C. 1994. Soil Analysis of the Glorieta Mesa Rock Art Site (AR-03-10-05147/LA90032). Letter report from John C. Phillips to Brent Abel dated March 23, 1994. Price, Steven D. 2006. The Quotable Horse Lover. New York: The Lyons Press. Quinn, Harry. 1989. 5LA5781: A Dated Rock Art Panel, Las Animas County, Colorado. In Rock Art of the Western Canyons, ed. Jane Day, Paul Friedman, and Marcia Tate, 27–29. Publication of the Denver Museum of Natural History and the Colorado Archaeological Society, Denver. Radin, Paul. 1916. The Winnebago Tribe. Bureau of American Ethnology Annual Report 37. Washington, DC: Smithsonian Institution. Reed, Alan, and Jonathan Horn. 1995. Cultural Resource Inventory of a Portion of the Picket Wire Canyonlands, Comanche National Grassland, Las Animas and Otero Counties, Colorado. Alpine Archaeological Consultants, Montrose, Colorado. Report on file with Comanche Grasslands, La Junta, CO. Renaud, Etienne. 1931a. Petroglyphs of Southeastern Colorado. In Archaeological Survey of Eastern Colorado, First Report, 64–87. Denver: University of Denver. ——. 1931b. A Unique Pictograph for Southeastern Colorado. El Palacio 30 (11–12): 175–78.

REFERENCES

241

——. 1936. Pictographs and Petroglyphs of the High Western Plains. Archaeological Survey of the High Western Plains, 8th Report. Denver: University of Denver. Ricks, M. F. 1995. A Survey and Analysis of Prehistoric Rock Art of the Warner Valley Region, Lake County, Oregon. Ph.D. thesis, Department of Systems Science, Portland State University, Portland. Ritter, Eric. 1994. Scratched Rock Art Complexes in the Desert West: Symbols for Socio-Religious Communication. In New Light on Old Art: Recent Advances in Hunter-Gatherer Rock Art Research, ed. David S. Whitley and Lawrence L. Loendorf, 51–66. Monograph 36. Los Angeles: UCLA Cotsen Institute of Archaeology. Robertson, Jean, and Nancy Robertson. 1975. Rock Art of the Raton Section of the Great Plains Province. American Indian Rock Art, vol. 1. Farmington, NM: American Rock Art Research Association. Rollings, W. H. 1989. The Comanche. New York: Chelsea House Publishers. Schaafsma, Curtis. 2002. Apaches de Navajo: Seventeenth-Century Navajos in the Chama Valley of New Mexico. Salt Lake City: University of Utah Press. Schaafsma, Polly. 1980. Indian Rock Art of the Southwest. Santa Fe and Albuquerque: School of American Research and the University of New Mexico Press. ——. 1992. Rock Art in New Mexico. Albuquerque: University of New Mexico Press. Schambach, Frank. 1977. Deer Ceremonialism in the Early Caddo Culture in Southwestern Arkansas. In Arkansas Archeology, ed. N. L. Trobowitz and M. D. Jeter, 132–97. Arkansas Archaeological Survey Research Series 15. Fayettville, AR. Schiavitti, Vincent, Lawrence L. Loendorf, and Erica Hill. 1999. Archaeological Investigations at Eleven Sites of Welsh Canyon in the Piñon Canyon Maneuver Site, Las Animas County, CO. Report on file with the Cultural Resource Program, Fort Carson. Scott, Sarah, and H. Edwin Jackson. 1998. Early Caddo Ritual and Patterns of Animal Use: An Analysis of Faunal Remains from the Crenshaw Site (3MI16), Southwestern Arkansas. The Arkansas Archeologist (37):1–38. Scott-Cummings, Linda. 1990. Results of Pollen Analysis from Site 5LA5586. Manuscript on file with PaleoResearch Laboratories, Denver. Scriver, Robert. 1990. The Blackfeet: Artists of the Northern Plains. Kansas City, MO: The Lowell Press, Inc. Secoy, Frank. 1953. Changing Military Pattern of the Great Plains. American Ethnological Society Monograph 21. Seattle, WA. Spock, Benjamin, and Stephen J. Parker. 1998. Baby and Child Care, 7th edition. New York: Pocket Books, a division of Simon & Schuster, Inc. Shimkin, Demetri. 1986. Introduction of the Horse. In Handbook of North American Indians, Vol. 11: Great Basin, ed. Warren L. D’Azevedo, 517–24. Washington, DC: Smithsonian Institution. Steed, Paul. 1984. Rock Art at Los Esteros. Artifact 22(3):1–89. Steinbring, Jack. 2003. Current Research on Wisconsin Petroforms. Electronic document, available at http://webpages.charter.net/magf/current%20research.htm. Steinbring, Jack, Jeffrey Behm, and Herman Bender. 2003. Petroforms in the North American Mid-Continent: A Neglected Phenomenon. In American Indian Rock Art, vol. 29, ed. Alanah Woody and Joseph T. O’Conner, 111–20. Tucson, AZ: American Rock Art Research Association.

242

L AWRENCE L. LOENDORF — THUNDER

AND

HERDS

Stevenson, Matilda Coxe. 1904. The Zuni Indians: Their Mythology, Esoteric Fraternities, and Ceremonies. Twenty-Third Annual Report of the Bureau of Ethnology for the Years 1901–1902, 8–634. Washington, DC. Stoffle, Richard, Michael Evans, M. Nieves Zedeno, Brent Stoffle, and Cindy Kessel. 1994. American Indians and Fajada Butte. Report prepared for the National Park Service. Tucson: Bureau of Applied Research in Anthropology, University of Arizona. Sundstrom, Linea. 1989. Archaic Hunting Practices Depicted in a Northwestern Plains Rock Art Style. Plains Anthropologist 34(124):149–69. ——. 1990. Rock Art of the Southern Black Hills: A Contextual Approach. New York and London: Garland Publishing Inc. ——. 2002. Steel Awls for Stone Age Plainswomen: Rock Art, Women’s Religion and the Hide Trade on the Northern Plains. Plains Anthropologist 47:99–119. ——. 2004. Storied Stone: Indian Rock Art of the Black Hills Country. Norman: University of Oklahoma Press. ——. 2006. Reading between the Lines: Ethnographic Sources and Rock Art Interpretation Approaches to Ethnography and Rock Art. In Talking with the Past: The Ethnography of Rock Art, ed. James D. Keyser, George Poetschat, and Michael W. Taylor, 49–72. Oregon Archaeological Society Publication, Portland, OR. Taçon, Paul S. C., and Christopher Chippindale. 1998. An Archaeology of Rock-Art through Informed and Formal Methods. The Archaeology of Rock-Art, ed. Christopher Chippindale and Paul Taçon, 1–10. Cambridge: Cambridge University Press. Tatum, Robert. 1944. The Petroglyphs of Southeastern Colorado. Southwestern Lore (3):38–43. Teit, James. 1906. The Lillooet Indians. Memoirs of the American Museum of Natural History 2, pt. 5:193–300. Thomas, Alfred B. 1929. An Eighteenth-Century Comanche Document. American Anthropologist 31 (2):289–98. Towner, Ronald. 2003. Defending the Dine’tah: Pueblitos in the Ancestral Navajo Homeland. Salt Lake City: University of Utah Press. Tratebas, Alice. 1993. Stylistic Chronology versus Absolute Dates for Early Hunting Style Rock Art on the North American Plains. In Rock Art Studies: The Post-Stylistic Era or Where Do We Go From Here?, ed. Paul Bahn and Michel Lorblanchet, 163–77. Oxford: Oxbow Books. ——. 2000. Evidence of Paleo-Indian and Archaic Hunting Techniques. American Indian Rock Art, vol. 24, ed. Frank G. Bock, 65–76. Tucson, AZ: American Rock Art Research Association. Trimble, Donald. 1980. The Geologic Story of the Great Plains. Geological Survey Bulletin 1483. Washington, DC: Government Printing Office. Turpin, Solveig, Herbert H. Eling, Jr., and Moises Valadez Moreno. 1996. The Mobiliary Art of Boca de Potrerillos Nuevo Leon, Mexico. Plains Anthropologist 41(156): 105–16. Wedel, Waldo R. 1961. Prehistoric Man on the Great Plains. Norman: University of Oklahoma Press. ——. 1967. The Council Circles of Central Kansas: Were They Solstice Registers? American Antiquity 32(1):54–63.

REFERENCES

243

Wellmann, Klaus. 1979. A Survey of North American Indian Rock Art. Graz, Austria: Akademische Druck und Verlagsanstalt. Wells, Nova. 1996. Petroglyphs of Saline River Valley, Kansas. Monograph 2. El Toro, CA: American Rock Art Research Association. Weltfish, Gene. 1965. The Lost Universe. New York: Basic Books. Whitley, David S. 1982. The Study of North American Rock Art: A Case Study from South-Central California. Ph.D. dissertation, University of California, Los Angeles. ——. 1999. A Possible Pleistocene Camelid Petroglyph from the Mojave Desert, California. San Bernardino County Museum Quarterly 46(3):107–8. ——. 2000. The Art of the Shaman: Rock Art of California. Salt Lake City: University of Utah Press. ——. 2008. Cave Paintings and the Human Spirit: The Origin of Creativity and Belief. New York: Prometheus Books. Whitley, David S., and Ronald I. Dorn. 1993. New Perspectives on the Clovis versus PreClovis Controversy. American Antiquity 58(4):626–47. Whitley, David S., and Lawrence L. Loendorf. 2005. Rock Art Analysis. In Handbook of Archaeological Methods, vol. II, ed. Herbert Maschner and Christopher Chippindale, 919–73. Lanham, New York, Toronto, and Oxford: AltaMira Press. Wilson, Ruby E. 1984. Frank J. North: Pawnee Scout Commander and Pioneer. Athens, OH: Swallow Press. Wintcher, Amanda. 2005. Rock Art and Landscape in the Piñon Canyon Maneuver Site, Southeast Colorado. In Making Marks: Graduate Studies in Rock Art Research at the New Millennium, ed. Jennifer Huang and Elisabeth Culley, 163–81. Occasional Paper 5. Tucson, AZ: American Rock Art Research Association. Winter, Joseph. 1988. Stone Circles, Ancient Forts, and Other Antiquities of the Dry Cimarron Valley. Santa Fe: New Mexico Historic Preservation Program. Wood, W. Raymond. 1962. A Stylistic and Historical Analysis of Shoulder Patterns on Plains Indian Pottery. American Antiquity 28(1):25–40. Wormington, H. Marie, and Richard Forbis. 1965. An Introduction to the Archaeology of Alberta, Canada. Denver: Denver Museum of Natural History. Wylie, Alison. 1989. Archaeological Cables and Tacking: The Implications of Practice for Bernstein’s Options beyond Objectivism and Relativism. Philosophy of Science 19: 1–18. Yazzie, Ethelou. 1971. Navajo History, vol. 1. Written under the direction of the Navajo Curriculum Center. Rough Rock Demonstration School, Many Farms, AZ. Chinle, AZ: Navajo Community College Press. Young Jane. 1988. Signs from the Ancestors. Albuquerque: University of New Mexico Press. Zeilik, Michael L. 1985. The Fajada Butte Solar Marker: A Reevaluation. Science 228 (4705):1311–13. Zier, Christian. 1999a. History of Archaeological Investigations. In Colorado Prehistory: A Context for the Arkansas River Basin, ed. Christian Zier and Stephen Kalasz, 25–42. Denver: Colorado Council of Professional Archaeologists. ——. 1999b. Paleoindian Stage. In Colorado Prehistory: A Context for the Arkansas River Basin, ed. Christian Zier and Stephen Kalasz, 73–99. Denver: Colorado Council of Professional Archaeologists.

INDEX

NOTE: Page numbers in italics indicate illustrations, photographs, and drawings. Abbot, P.O., 47 Abel, Brent, 57–58 abstract, defined, 45 Abstract Curvilinear, 62 abstract petroglyphs: analyzing differences in, 63; in Archaic Period, 36, 48–53, 61–63, 67, 69, 70, 74–75, 223; at Crack Cave, 208; defined, 48, 147; effects of consciousness on, 78. See also altered states of consciousness; in Paleo-Indian Period, 77–79; in Pecked-I style, 91–92, 105; in Pecked-II style, 123 Abstract Rectilinear, 62 Agenbroad, Larry, 73 Algonkian-speaking groups, 214, 225 altered states of consciousness, 78, 102–103, 120, 136–140, 151, 182. See also spiritual life altithermal, 64–65, 67, 68 Ancient Hogback site, 74–75, 77–79 Anderson, David, 87 Andresen, Ken, 28 animal cults, 158 animal images. See also quadruped figures; specific animal names: in Apache rock art, 178; in Archaic Period, 62–63, 77, 79; death, in rock art, 96, 97–98; in oversize rock art, 193–198; in Painted rock art style, 146; in Pecked-I style, 92, 101, 105;

in Pecked-II style, 116, 131, 133; penetrated by spears, 127, 133, 157, 229 antelope, 131, 213 anthropomorphic images: action-oriented, 124, 125, 146–147, 222; in Apache rock art, 180–181; in Archaic Period, 62–63, 69, 71, 221; assigning to time periods, 37; in Bear Dance Ceremony, 161; with bears, 156–157; with bow, 133; clown dancers, 159–160, 161, 181–183; dancing figures, 124, 125, 139–140, 157, 158–162, 221, 223, 228; in Diversification Period, 133; drawings of, 69, 72, 100, 177; effigies, 122, 127–128; female figures, 119, 199–201, 217–218; and footprint imagery, 99–101, 104–105; gans, 177, 181–183, 186, 228; and hand imagery, 94–95, 185; in headdress, 156–157; in Historic Period, 198, 199–200, 213; horned, 171, 172, 181; on horseback. See horse-andrider motif; oversize figures, 193–198, 196; in Painted rock art style, 143, 145–146, 152–153; in Pecked-I style, 81, 85, 86, 88, 91, 92, 95–98, 99–101; in Pecked-II style, 114, 116, 124–127, 131–134; profile view, 125, 131, 157, 185, 199; in Protohistoric Period, 171, 174, 186; red-painted, 115; seriation of, 35; with smallpox, 199–200,

244

INDEX 200, 213–214; spear carrying, 156; transformation to bird, 137–138, 151, 196, 213; transformation to deer, 139–140; two-lance figure, 202–203, 203; V-necked, 202–203; antlers: branching, 38–39, 54–55, 70, 95–98, 106–107, 146; interlocking, 114, 114; net-like, 101–104, 102, 138; in Pecked-I style, 85, 88–89; for tool-making, 118, 216 Apache mountain spirits, 181, 183 Apache peoples: ceremonies of, 183; gans dancer, 181–183, 186, 228; heart lines motif, 173, 178; horses and, 191; migration of, 164–167, 227–228; in Painted rock art style, 180–181, 228, 229; in Protohistoric Period, 32, 173–178 Apishapa culture: in Painted rock art style, 157, 162; in Pecked-II style, 110–112, 125, 131–141; in Protohistoric Period, 166; rock art of, 224, 229–231 Apishapa River, 27, 110–111 archaic, defined, 45 Archaic Period: abstract petroglyphs in, 36, 48–53, 61–63, 67, 69, 70, 74–75, 223; chronological sequences, 70; Early Archaic, 64–68; geography of, 62; Late Archaic, 70–71; Middle Archaic, 68–69; pecked petroglyphs, 209–210; rock art, 62–71, 221–223; seriation of, 36 Arrow Canyon, 136 arrow points. See projectile points astronomical features, 122, 207–208 Athapaskan-speaking groups: Castle Garden style, 183–184; migration of, 165–167, 226–228; rock art, 167, 173; turtles as shields, 179 atlatls, 75, 114, 134, 148 Baker, Galen, 27 basalt formations, 15, 23–24, 24, 41, 51–52, 52, 123

245

bas-relief, 185, 186, 186 Battiste Good drawings, 199, 214 Baumhoff, Martin, 34, 62 Bear Dance site, 139, 154–161, 155–157 bear images: in black pigment, 205; in cinnamon pigment, 213; in Historic Period, 156–157, 205–206; oversize figures, 194, 197–198, 213; in Painted rock art style, 159–160, 194–195; in Pecked-I style, 81; in Pecked-II style, 131; penetrated by spears, 157; in red pigment, 195, 198 bear medicine, 183 bear tracks, 61, 223, 224 bedrock metates, 29, 82, 107–108, 117–120, 118, 123, 145, 154–156, 161 Benedict, James, 64–66 Bent Canyon, 69, 70, 71, 125 Big Hands Hunter site, 94–103, 96–98, 100, 102, 222 Big Horn Medicine Wheel, 122 bighorn sheep, 77, 81, 221 biographic tradition, 208–212 bird motifs: at Corral site, 126–127; in Late Archaic Period, 70–71; in Pawnee lifeways, 197; pecked and painted, 147 bird tracks, 64, 66, 104–105 birthing scene, 105 bison: in Archaic Period, 68, 73, 221; in black pigment, 205; in Developmental Period, 88–90; drawings of, 174, 179; with heart line, 179; oversize figures, 213; in PaleoIndian stage, 31; in Pecked-I style, 81; in Protohistoric Period, 133, 171, 172, 178; Black Hills, South Dakota, 28, 62, 64, 68, 77 Blakeslee, Donald J., 215 bleeding. See blood and bleeding images blinder syndrome, 16, 17 blood and bleeding images, 96

246

L AWRENCE L. LOENDORF — THUNDER

Blue Bull Cave, 52 Boca de Potrerillos site, 78 Bock, A.J., 58–59 Bock, Frank G., 58–59 boulder effigies, 122, 127–128 bow, in rock art, 55, 180, 180, 185 Breternitz, David, 27 Buckles, William G., 27–28, 47–48, 199, 213 buffalo, 214–218 Bull Pasture site, 82–86, 84, 86, 103 Burnard-Hogarth, Linda, 226 Burton, Robert J., 27–28 Caddoan deer-hunting rituals, 230 Caddoan-speaking peoples, 111, 158–162, 166–167, 208, 225 cairns, 144 calendar, 207 “calling station”, 127 Campbell, Robert, 27 Canyon de Chelly, 52 Carved Rock site, 108 Casteet, James W., 46 Castle Garden style, 183–184, 226 cation-ratio dating (CR), 39–40, 42, 124–125 cedar trees, 159 cedar wreath, 160–161 Celtic voyagers, 206–207 Central High Plains. See High Plains ceramics, 93, 111, 119 cervid species. See deer figures Chacuaco Canyon, 21, 25, 111 charcoal pigment dating, 60, 193 Chase, Haldon (Hal), 26–27 Chauvet cave, 94–95 Cheyenne peoples, 32, 192, 212 Chippindale, Christopher, 42, 221 Chomko, Stephen, 87–88 Cimarron phase, 223 cinnamon color, in pictographs, 194, 213 circle motif. See also concentric circle motif: in Archaic Period, 61, 68; bisected, 67, 116; double-outlined, 48; interconnected, 95; in Painted rock art

AND

HERDS

style, 150–151; in Pecked-I style, 81, 85, 90, 105; smallpox as, 199–200 classification, in Pecked-I style, 105; polythetic system of, 39; in rock art research, 33–34 Clay Creek, 46, 47, 47–50, 49, 68, 69 climate. See also weather: in Apishapa phase, 110–111; in central High Plains, 24–25, 32; Early Archaic, 64–66; effects on varnish, 41; and human behaviors, 121; at Piñon Canyon, 51 Clovis Period, 31, 72–73, 77 clown dancers, 159–160, 161, 181–183 Cole, Sally, 29, 63–64, 81, 142, 163 Colorado Piedmont, 22–23 Columbia Plateau, 229 Comanche National Grasslands, 12, 29, 30, 198 Comanche peoples, 32, 166–167, 192, 203–204, 211–213 communal hunting, 113, 127–129, 135–136, 140–141, 144–145, 221 concentric circle motif, 48, 54, 61, 64, 67, 69, 224 Conchas Reservoir, 67 Conkey, Margaret, 225 Conway, Thor and Julie, 136 Cooper site, 74 Corral site, 126–131, 127, 128 Cortés, Hernán, 191 Crack Cave, 206–208 Crashing Thunder (Wisconsin Winnebago Indian), 135 Crenshaw site, 230 Cross Ranch Animal Drive, 128 Crow Indians, 176 cultigens, 111 cupules, 117–118, 120, 200–202, 201 curvilinear styles, 34, 35, 36, 37, 62, 70, 151 Dakota sandstone, 47, 47, 113, 168 dancing figures, 124, 125, 139–140, 157, 158–162, 221, 223, 228

INDEX dating techniques, 33–34, 39–42, 43, 60, 77, 157 deer figures. See also antlers: in Archaic Period, 221; drawings of, 174; hunting rituals, 230; in Painted rock art style, 26; in Pecked-I style, 81, 84–85, 88, 95–98, 101, 103–104; in Pecked-II style, 113–114, 131; at Renaud rockshelter, 163 Developmental Period, 32; in Apishapa culture, 110, 223; characteristics of, 40; hoof design in, 104; incised figures, 210; in Pecked-I style, 82, 100, 103; quadruped figures, 222 dew claws, 171, 178 dikes. See basalt formations Dinetah, 226–227 dinosaur footprints, 87, 198 Dinosaur National Monument, 27–28 Dinwoody rock art, 225 Dismal River Aspect, 167 Diversification Period: Apishapa culture in, 109–110, 157; characteristics of, 40, 124, 157; Painted rock art in, 162; Pecked-I petroglyphs in, 109, 110; Pecked-II petroglyphs in, 125, 133; seriation of, 36 Doctor Association, 158–160 Doctor Lodge, 158–159, 160–161 dogs, 81, 131–133, 152, 167, 170, 181, 185 Dondelinger, Norman W., 27 Dorn, Ronald I., 39, 41, 59, 60, 77 Dorsey, George A., 138–140 dots: in Archaic Period, 50, 78, 223; in Painted rock art style, 147, 151; in Pecked-I style, 96–97; in Pecked-II style, 116; in Protohistoric Period, 172, 184; seriation categories, 35 Double Woman, 217 eagle motif, 151, 162, 208 eagle-catching pits, 56 Early Archaic, 32, 64–68, 78, 221–222 Early Entry, High Plains Refugee Hypothesis Model, 227

247

Early Entry, Mountain Route Refugee Hypothesis Model, 166 effigies, 122, 127–128 El Cuartelejo, 167 elk, 103–104, 221 ethnography, 16–17, 221 Evening Star, 162 Ewers, John C., 163 eye-hand coordination, 94 eyes, weeping motif, 211 Faris, Peter, 29 faunal remains, 112, 170 female figures, 119, 199–201, 217–218 Feyhl, Kenneth, 216–217 Fidler, Peter, 216 Fire God, 199 floral remains, 170 flute players, 223, 224 Folsom Period, 31, 72–73 footprints: dinosaur, 87, 198; human, 61, 99–101, 100, 104–105, 193, 209 Francis, Julie E., 225 frequency seriation. See seriation Game Drive site, 127, 145, 145–151, 147 gans dancer, 177, 181–183, 182, 186, 228 Gault stones, 77 gender associations, 119–120, 162, 163, 199, 201, 217–218 geoglyphs, defined, 122 geometric designs, 117, 163, 174, 223 Gilpin, Dennis, 223 Glassow, Michael, 223 Glorieta Mesa, 56–61, 58–59, 65, 66, 67–68 Good Heart, 140, 230 Goodwin, Grenville, 182 graffiti. See vandalism green pigments, 184 grid-like forms: in Archaic Period, 61, 69–70, 79–81, 88; in Painted rock art style, 148–151; in Pecked-I style, 105–109; in Pecked-II style, 114; in Protohistoric Period, 174

248

L AWRENCE L. LOENDORF — THUNDER

Gunnerson, James, 167 guns imagery, 210 Gurnsey, B.H., 194, 195 Hackberry Springs, 28 hand imagery, 94–95, 185 Harmon, Daniel, 136 Hartley, Ralph, 29 Haury, Emil, 226 Hawken site, 68 headdress, 156–161, 180–185, 210, 212, 228 heart lines motif, 173, 178, 179 Heizer, Robert, 34, 62 hematite, 152, 215 Hesse, India, 73 Hicklin Springs site, 69, 196, 213 High Plains: climate, 24–25; cultural chronology, 25, 30–34; map of, 20; vs. Park Plateau, 223–226; physical environment, 20–24, 21; High Plains Archaeological Expedition, 25–30 Historic Period: biographic tradition, 208–212; chronological sequences, 210–211; Crack Cave, 206–208; incised rock art, 192–193; oversize figures, 193–198; Picture Canyon, 198–208; techniques and forms, 212–214 Hogback 3 site, 106–107 Hogup Cave, 78 horse motif, 29, 79, 190–193, 205, 209–213 horse-and-rider motif, 203–204, 204, 209, 211, 211, 211, 212 human effigies, 122, 127–128 human footprints, 61, 99–101, 100, 104–105, 193, 209 human images. See anthropomorphic images human-bird images, 137–138, 151, 196, 213 hunter-gatherer, 56, 64–65, 68, 166 hunting blinds, 175

AND

HERDS

hunting scenes, 120, 132–134, 133–138, 148–151, 221, 231 hunting techniques. See also communal hunting: in Archaic Period, 82, 85, 221–222; and Athapaskans, 180; dogs used in, 221; with horses, 191; hunting blinds, 175; in Pecked-I style, 98, 101–102; in Pecked-II style, 127–129, 132–133; in prayer stations, 93–94; pre-hunt rituals, 140; rock art as instruction, 136–137, 150; weapons used in, 203 Ice Age, 32, 51, 72 iconographic approach, 16 incised figures. See also abstract petroglyphs; pecked figures: in Archaic Period, 75–76; in biographic tradition, 209–210; in Historic Period, 198, 211–212; horses, 192–193; in Paleo-Indian Period, 74–76, 78, 79; temporal patterning, 77 incised lines, 118, 202, 207, 209 iniskims, 214 James Peak, 68 Johanson, Donald, 86–87 Johnson, Lyndon, 46 Kalasz, Stephen, 30 Keyser, James D., 205–206, 208–209, 229 Kiowa peoples, 32 Kutenai Indians, 216 Late Archaic Period, 32, 36, 40, 70–71, 82, 222 Late Entry, High Plains Refugee Hypothesis Model, 166–167 Late Prehistoric stage, 32, 36 Lee, Laurie, 74 Legend Rock site, 77, 221–222 linear motifs, 63, 71 Liu, Tanziu, 77 llama, 77

INDEX lollipop motif, 48, 53, 67 Long Lake site, 78 loop-line figure, 90 L-shaped feet, 85, 103, 104, 146, 157, 201–202 Magne, Martin, 226–227 mammoths, 31, 73–74, 77, 79 manos and metates. See also bedrock metates: in Archaic Period, 32, 59, 82; in Developmental Period, 93; at Painted rock art sites, 152; at Pecked-I petroglyph sites, 107–108; at Pecked-II petroglyph sites, 117–118; use of, 119 Maselli, Michael, 199, 213 Massacre Bench site, 78 Matson, R.G., 226–227 McGlone, William (Bill), 218, 221 McNees Crossing, 163–164, 228 Medicine Creek Cave, 28 medicine, animal, 139–140, 183, 225 medicine bundle, 135, 208, 214 Medicine Men dance, 139 medicine wheel, 122, 207 menstruation ritual, 217 mescal beans, 139 metates. See manos and metates Middle Archaic, 32, 36, 68–69 Mitchell, Mark, 29, 30, 213 mobiliary art study, 78 Mobley, Charles, 63 Mojave Desert, 77 Morning Star, 162 Morris, Earl, 226 Mount Albion Complex, 68 Mount Churchill, 166 mountain lions, 101 Mountain Route model, 166–167 mountain spirits, 176, 177, 178, 181, 183 Murie, James R., 160 Navajo peoples, 165–166, 178–180, 226–227 nets and snares: in Painted rock art style, 150, 150; in Pecked-I style, 81, 85,

249

95, 101–102, 109; in Pecked-II style, 114, 115, 116, 126 North, Frank, 140 Nuevo Leon, Mexico, 78 O’Brien, Patricia, 162, 208 Ocate Micaceous ceramics, 187 Occam’s Razor, 207 Ogam language, 206–208 Olsen-Chubbuck site, 73 Opler, Morris E., 176 out-of-body experience. See consciousness, altered states of oversize figures, 193–198, 196, 213 overstriking, 231 Owens, Mark, 108, 130 pain, represented in rock art, 200, 213–214 Paleo-Indian Period, 31–32, 72–79, 222 Palmer Divide, 23 panels, defined, 33 parfleches, 163 Park Plateau, 223–226 Passports in Time (PIT), 58–59 Pawnee Bear Ceremony, 158–161 Pawnee peoples: Bear Dance ceremony, 158, 160–162; cosmology, 208; and deer shamans, 139; horses and, 191; migration of, 32, 111, 112; transformation, 213 pecked figures. abstract, 76, 223; in Archaic Period, 47, 50, 54–55, 62, 68; in biographic tradition, 209–210; defined, 52; geometric, 223; in Historic Period, 198; in Pecked-I style, 81, 88, 95, 103; in Pecked-II style, 114; representational figures, 79; strike marks, 116; vertical lines, 69 petroform, defined, 122 Petroform site, 121–125, 123–124, 173 petroglyphs. See also pictographs; rock art: action-oriented, 125; Apache, 181, 187; atlatls represented in, 148; chronological age of, 39–42; outlined-pecked, 181; shooting arrows

250

L AWRENCE L. LOENDORF — THUNDER

at, 135–136; solidly pecked, 170; stipple-pecking, 170, 172–173; strike marks, 116, 117, Phillips, John C., 59, 60 Philmont Scout Ranch, 223–225, 224 Picket Wire Canyon, 21, 21, 87, 91, 93, 185 Picnic site, 215 pictographs. See also petroglyphs; rock art: in Painted rock art style, 152; with red paint, 198; use of cinnamon color, 194, 213 Picture Canyon, 198–208, 201 pigments, use of: black, 205; cinnamon, 194, 213; green, 184; red, 55, 114, 117, 129–131, 143, 152, 163–164, 184, 193; white, 184; yellow, 129–131 Piney Creek alluvium, 47–48 Piñon Canyon Maneuver Site, physiography, 87; rock formations, 23–24; terrain, 151 Pistone site, 78 pit structures, 55, 55, 56, 177 Plains Apaches. See Apache peoples Plains Biographic, 223 Plano Period, 72–73 Point site, 91–92, 91–93 polythetic system of classification, 39 Ponil phase, 223 prayer stations, 93–94, 140–141, 176, 205, 214–215, 230 prehistoric site, defined, 32–33 projectile points: in Apishapa phase, 111; in Archaic Period, 59–60, 67, 68; corner-notched, 93, 152, 169; in Developmental Period, 82, 152; in Diversification Period, 152; in PaleoIndian stage, 31, 73; Scallorn, 93; side-notched, 170, 227; stemmed, 93; tool grooves, 118, 216; triangular arrow, 187; unnotched, 170, 187 pronghorn: in communal hunting scenes, 221; in Painted rock art style, 144–145; in Pecked-I style, 81,

AND

HERDS

103–104; in Pecked-II style, 112, 126–131, 140 protohistoric, defined, 165 Protohistoric Period: anthropomorphic images in, 171, 174, 186; Apache rock art in, 178–188, 206, 227; human development during, 165–167; Rio Grande Style, 173, 184–186, 227–228; shield-bearing warriors, 183–184; Stone Structure site, 173–177; Sue site, 168–173 Pueblo II Period, 223, 224 Pueblo peoples, 173, 184–186, 191, 199, 223, 227–228 Pueblo Revolt of 1680, 192 Purgatoire Bear, 194 Purgatoire Canyon, 21 Purgatoire Painted petroglyphs: and Apache peoples, 164; Bear Dance site, 154–162; bird motifs’ role, 197; chronological sequences, 162; color palette, 143; dating of, 157; Game Drive site, 143–151; geography of, 163; quadruped figures, 26; Rock Crossing site, 151–154; stylistic definition of, 26, 142, 228 Purgatoire Pecked-I petroglyphs: and Apishapa culture, 226; Archaic petroglyph sites vs., 106; Big Hands Hunter site, 94–103; cultural chronology, 82; dating of, 108–109; defined, 81, 131; in Developmental Period, 90–92; geography of, 105; pecked geometric, 223; quadruped figures, 84; at Zookeeper site, 86–94 Purgatoire Pecked-II petroglyphs: analysis of, 131–141; and Apache peoples, 164; and Apishapa culture, 110–112, 226; Corral site, 126–131; defined, 110–112, 131; pecked geometric, 223; Petroform site, 121–125; RedTail Rockshelter, 112–120; as shrines, 140; strike marks, 116, 117, 135–136

INDEX quadruped figures. See also antlers: in Apache rock art, 178, 205–206; in Archaic Period, 38, 54–55, 69, 70, 71, 222; assigning to time periods, 36–38, 40; at Big Hands Hunter site, 95–98, 97, 98, 102; at Bull Pasture site, 84; chronological sequences, 108–109; classification of, 106–107; Developmental Period, 222; Diversification Period, 126; drawings of, 53, 54, 71, 96–99, 146–147, 187; fully pecked, 174; hoof treatment, 39; oversize, 193–198; in Painted rock art style, 143, 145, 146, 146–147, 152–153; in Pecked-I style, 81, 84–85, 86, 88, 95, 104; in Pecked-II style, 113–115, 124–125, 130, 135; properties of, 38–39; ribstones, 205; seriation of, 35; strike marks, 116, 117, 135–136; U-shaped hooves, 39, 103, 171, 172; at Zookeeper site, 89–90 quail, 162, 208 Radin, Paul, 135 radiocarbon dating, 42 rattle, 75, 75, 124 rectilinear styles, 34–37, 62 Red Guns site, 210–212 red pigment, 55, 114, 117, 129–131, 143, 152, 163–164, 184, 193 Red-Tail Rockshelter, 112–120, 113, 115–118, 121, 163 Renaud, Etienne B., 26, 26, 198–199, 205 Renaud rockshelter, 163 Renaud’s Shelter, 153, 163 representational figures, 79 ribstones, 205, 205, 214–218, 215 Rio Grande Style, 173, 184–186, 185–186, 227–228 Ritter, Eric, 78, 79 rituals, 94, 119, 140, 208, 217, 229–230 Roaming Chief (Pawnee), 160 Robertson, Nancy, 28–29, 83

251

rock art: assemblage, 65; avocational studies of, 28; chronological sequences, 27–28, 29; classification of, 27; cultural chronology, 30–32; dating techniques, 33–34, 39–42; defined, 19; panels, 33; scientific method to, 28; study of, 45; temporal and spatial frameworks, 29; typology, 63, 64; water and, 50–51, 67 rock art research: access to sites, 83; ambiguous features and, 55–56; with auger probes, 129, 130; chronology, 33–36, 70; classification, 33–34; ethnic identity, 180; history of, 25–30; Occam’s Razor, 207; rules of, 60–61; scientific method to, 28; seriation categories, 34–36; significance of, 50; style, defined, 33; surveying, 154; tool grooves, 216–217; working on private land, 83; x-ray fluorescence analysis, 130–131 Rock Crossing site, 151–154, 153, 163 Rodgers, Kendra, 156 Rongstad, Orrin, 87 Rudolph, John and Daphne, 28 Scabby-Bull (Skidi Pawnee), 139–140 Schaafsma, Polly, 62 Schambach, Frank, 230 Scott, Glen, 47 Scottsbluff Period, 31 scroll symbol, 213–214 Segesser, Philip von Brunegg, 167 Segesser I and II, 167, 168 seriation, 34–36, 37 shamans: Apache, 182–183; in Bear Dance Ceremony, 159–162; hunting techniques of, 136–141, 144, 229; Pawnee, 197, 208, 215; as teachers, 159–160 shield-bearing warrior, 86, 167, 181, 183–185, 225, 226 shields, 85–86, 168, 183–186. See also shield-bearing warrior Shoshone groups, 191–192

252

L AWRENCE L. LOENDORF — THUNDER

Shulawitsi, 199 smallpox, 199–200, 200, 213–214 Smith site, 176, 177, 187 Snake Blakeslee site, 27 snake figures, 90, 104 snares. See nets and snares Sorenson site, 93 Spain: horses and, 191–192; observation of Native peoples, 166, 167, 203, 230; and Pueblo Revolt of 1680, 192 spatial clustering, 106, 107, 120, 121 spears and spear figures: in Historic Period, 202–203; in Painted rock art style, 148, 156–157, 161; in Pecked-I style, 95, 98, 99, 100, 101–102; in PeckedII style, 114, 119, 127, 133–134; symbolic killing, 101–102, 132–134 spirals, 63–64, 95 spiritual life: Apache peoples, 176; at Big Hands Hunter site, 96; of Pawnee peoples, 158; Plains Indian tribes, 197–198; at Zookeeper site, 94 Star site, 205, 215 Star Tail Quadruped, 105 starburst, 61, 117, 121 Steed, Paul, 63 Stegler, Robert, 25–30 Steinbring, Jack, 122 Stevenson, Matilda Coxe, 135 Stone Structure site, 173–179, 174, 186 stones, as human effigies, 122, 127–128 strike marks, 116, 117, 135–136 Sue site, 168–173, 169, 171–172 sun, in rock art, 206–207 Sun Temple site, 206 sunbursts, 64 Sundstrom, Linea, 62, 64, 217 Taçon, Paul S.C., 42, 221 Tatum, Robert M., 27 territorial boundaries, 225–226 Thomas, Alfred B., 203 thunderstorms, 14–15, 23–24, 25, 122 Tie Creek site, 122

AND

HERDS

Tierra Blanca phase, 227 Tolar site, 204 tool grooves, 118, 205–206, 216–218 tools. See also hunting techniques: abrading, 183; edge-ground, 108; gender associations, 119–120, 217; stone, 144, 169, 170, 206 Towner, Ronald, 166–167, 227 tracks. See bird tracks; footprints trade network, 192 trance experiences. See consciousness, altered states of transformation: human to bird, 137–138, 151, 196, 197, 213; human to deer, 139–140 turkeys, 153, 153, 162–163, 215 Turpin, Solveig, 78 turtles, 122, 171, 178–179, 179 two-lance figure, 202–203, 203 Uncompahgre Plateau, 28, 64 “Ursa Major”, 194 U-shaped figures, 55, 88, 104, 146 U-shaped hooves, 39, 103, 171–172, 172, 174 Ute Indians, 28, 191–192 Uto-Aztecan-speaking groups, 225 Valley of the Shields site, 183 vandalism, 118–119, 195 varnish micro-lamination dating (VML), 39, 41, 42 Vawser, Anne Wolley, 29 vegetation. See floral remains vision quests, 136–137, 175–176, 182. See also altered states of consciousness VML dating. See varnish micro-lamination dating (VML) V-shaped figures, 61, 146, 153 vulvar motifs, 201, 217 Wagon Mound site, 67, 105 water: effects on rock art, 50–51, 64–67; petroglyph locations to, 64–66 water serpent, 159 Waugh, Mike, 74–75, 221

INDEX weaponry, 98, 136, 167, 171, 193–194, 203, 210 weather, effects of, 46, 114, 120. See also climate weathering rind organic dating (WRO), 39, 41 Wedel, Waldo R., 220 weeping eyes motif, 211 Weltfish, Gene, 158–159, 160 white pigment, 184 White River tephra, 166 Whitfield, Anne, 30, 30, 199, 213 Whitley, David S., 77, 137, 229 Wichita deer dance, 139 Wichita peoples, 111, 139–140, 192, 207 Wilcox, Waldo, 83–84 Winter, Joseph, 28–29 Wissler, Clark, 160

253

witch woman, 159 Witt, C.C., 207–208 women, presence of, 119, 168, 199–201, 217–218. See also gender associations women’s fertility ceremonies, 120 WRO dating. See weathering rind organic dating (WRO) Wylie, Alison, 42–43 x-ray fluorescence analysis, 130–131 yellow pigment, 129–130, 129–131 Yellowbird (Skidi Pawnee), 138–139 Zier, Christian, 30 zigzag patterns, 61–63, 74–75 Zookeeper site, 36, 86–94, 89–90, 103, 105

ABOUT THE AUTHOR

awrence Loendorf is an anthropologist and archaeologist whose research focuses on the North American intermountain West, ethnography, traditional cultural properties, and rock art. He currently directs a project to document the petroglyph sites on the hogback at the Piñon Canyon Maneuver Site in southeastern Colorado. Recent books include Ancient Visions: Petroglyphs and Pictographs of the Wind River and Bighorn Country, Wyoming and Montana, University of Utah Press (with Julie Francis); Restoring a Presence: American Indians and Yellowstone National Park, University of Oklahoma Press (with Peter Nabokov); and Mountain Spirit: The Sheep Eater Indians of Yellowstone, University of Utah Press (with Nancy Medaris Stone).

L

254